Error compiling certain uses of std::tuple in Solaris Studio 12.4 beta July refresh

Whilst attempting to compile Boost 1.54 using the Solaris Studio 12.4 beta July refresh with -std=c++11 I get an error building libs/regex/src/cregex.cpp that looks like it might be due to a problem with the compiler. (clang++ on Mac OS X and g++ on Linux can certainly build the same file without problems - I can post more details about them if you think this is a problem in the Boost code rather than the compiler.)
The error from Solaris Studio is:
Error: Could not find a match for std::_Tuple_impl<0, int&&>::_Tuple_impl(std::tuple<int&&>, int) needed in std::tuple<int&&>::tuple<int, void>(int&&).
"/opt/SolarisStudio12.4-beta_jul14-solaris-x86/lib/compilers/CC-gcc/include/c++/4.8.2/tuple", line 868:     Where: While instantiating "std::tuple<int&&>::tuple<int, void>(int&&)".
"/opt/SolarisStudio12.4-beta_jul14-solaris-x86/lib/compilers/CC-gcc/include/c++/4.8.2/tuple", line 868:     Where: Instantiated from std::forward_as_tuple<int>(int&&).
"/opt/SolarisStudio12.4-beta_jul14-solaris-x86/lib/compilers/CC-gcc/include/c++/4.8.2/bits/stl_map.h", line 485:     Where: Instantiated from non-template code.
1 Error(s) detected.
To reproduce this you can download the pre-processed source code from http://pastebin.com/jtsx1k79 and paste it into a file called cregex.pre.cpp. Then run:
CC -std=c++11 -mt -m64 -c -o cregex.o cregex.pre.cpp
to get the error I saw.
In case it's relevant, I'm working on Oracle Solaris 10 1/13 s10x_u11wos_24a X86.

This looks very much like a recent bug with std::map introduced with the change to g++ 4.8 headers in July Beta:
19159587 C++11: errors on a simple <map> usage
It has been reported on forums - +d flag can cause C++ compilation failure in 12.4 beta refresh
This bug will be fixed in a final release.
Unfortunately there is no workaround for this problem.
regards,
Fedor.

Similar Messages

Error Compiling when using RGB Color Corrector and Unsharp Mask

If I apply the RGB Color Corrector and and Unsharp mask to a clip within a sequence and try to render the sequence or a small portion of the work area I get the Error Compiling: Unknown error.
This is occurring within a project I have updated to Premiere CS6 from Premiere CS5.
I have tried creating a new sequence and importing the same clip and other clips. Anytime I try to use both the Unsharp Mask and RGB Color Corrector I am unable to render. I have tried this with Mercury Playback both in hardware and software modes with the same result.
Has anyone else seen something like this?

Okay, glad this occurs for others too and I am not experiencing something isolated. The issue here is this shouldn't be happening at all.
Premiere CS5 has no problem with this.
Yeah if I turn off one effect or the other no problem. Unfortunately it doesn't matter the order I place the effects in I get the same error regardless of which is placed first.
Nesting does work. If I create a sequence, apply one or the other to the clip in the sequence then embed the sequence in another sequence and apply the remaining filter to the sequence containing the clip then I can get my color correction and edge definition. But this is a work around.
There seems to a bug here since there were no issues when using RGB Color Corrector and Unsharp Mask simultaneously applied to one clip in CS5. I doubt there will be a fix for this anytime soon so it looks like I will have to use another method for color correcting or apply RGB color correction to clips already embedded in a sequence as I move forward into solely using CS6.
(How this can become an issue is that after applying effects and color correction to several clips within a sequence I like to render that entire sequence. Then this sequence is ebedded in another main sequence and I can add additional effects and transitions and overlays without killing the live playback speed and only have to render smaller amounts of content from additional tracks. It ends up saving a lot of time after the initial length of rendering.)
Thanks for the suggestions.

Solaris Studio 12.4 Beta unable to compile simple C++ program

The commands and output are shown below
cat test.cpp; /opt/oracle/solaris_studio/bin/CC -std=c++11 -o test test.cpp
#include <random>
int main () {}
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 41: Error: The function "__builtin_ia32_bsrsi" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 49: Error: The function "__builtin_bswap32" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 89: Error: The function "__builtin_ia32_rdpmc" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 97: Error: The function "__builtin_ia32_rdtsc" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 105: Error: The function "__builtin_ia32_rdtscp" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 113: Error: The function "__builtin_ia32_rolqi" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 121: Error: The function "__builtin_ia32_rolhi" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 137: Error: The function "__builtin_ia32_rorqi" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 145: Error: The function "__builtin_ia32_rorhi" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 161: Error: The function "__builtin_ia32_pause" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 178: Error: The function "__builtin_ia32_bsrdi" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//ia32intrin.h", line 186: Error: The function "__builtin_bswap64" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//adxintrin.h", line 36: Error: The function "__builtin_ia32_addcarryx_u32" must have a prototype.
"/opt/oracle/solaris_studio/lib/compilers/CC-gcc/lib/gcc/x86_64-unknown-linux-gnu/4.8.2/include//adxintrin.h", line 45: Error: The function "__builtin_ia32_addcarryx_u64" must have a prototype.
14 Error(s) detected.
I install the studio with the tarball file in CentOS 7. I know that officially it only support RHEL 5 & 6, but this is clearly not an OS issue

This is a known bug in 12.4 Beta:
18688928 C++11 <random> header includes x86intrin.h on Linux, leading to failures
This problem is fixed in a final 12.4 release, which will be announced really soon.
regards,
Fedor.

Use of __typeof__ in Solaris Studio 12.4

Solaris Studio 12.3 had a __typeof__ operator similar to decltype in C++11. This is available in a portable wrapper using the Boost macro BOOST_TYPEOF_TPL. It looks like the Solaris Studio team contributed the modification to Boost which added this in ticket 5745.
Now Solaris Studio 12.4 has a C++11 mode which supports decltype. However, is __typeof__ still supposed to work in Solaris Studio 12.4's C++03 mode? It doesn't seem to work as well as it used to.
The following program deduces the return type of operator() using __typeof__:
#include <boost/typeof/typeof.hpp>
#include <boost/type_traits/remove_reference.hpp>
#include <iostream>
#include <typeinfo>
// Helper to deduce the result type from a member function pointer.
template<typename T>
struct member_function_result_type
template<typename T, typename R, typename A1>
struct member_function_result_type<R (T::*)(A1) const>
    typedef typename boost::remove_reference<R>::type type;
template<typename T>
struct function_result_type
    typedef BOOST_TYPEOF_TPL(&T::operator()) F;
    typedef typename member_function_result_type<F>::type type;
template<typename R, typename A>
struct function_result_type<R (*)(A)>
    typedef typename boost::remove_reference<R>::type type;
struct my_functor
    float operator()(float numerator) const
        return numerator / 2.0f;
int main(int, char **)
    typedef typename function_result_type<my_functor>::type my_type;
    my_functor f;
    my_type ret = f(5.0f);
    if (typeid(my_type) == typeid(float))
        std::cout << "As expected " << ret << std::endl;
    else
        std::cout << "Oh dear " << ret << std::endl;
    return 0;
If you build this with Solaris Studio 12.3 as follows:
CC -m64 -I/usr/local/include/boost-1_54 main.cc
then it compiles and running a.out shows the return type of float has been correctly deduced.
However, with Solaris Studio 12.4 the same command:
CC -m64 -I/usr/local/include/boost-1_54 main.cc
results in:
"main.cc", line 23: Error: Unexpected type name "F" encountered.
"main.cc", line 42: Warning: "typename" must be used within a template.
1 Error(s) and 1 Warning(s) detected.
Now, obviously Solaris Studio 12.4 has a C++11 mode and it is possible to get this program to work by changing BOOST_TYPEOF_TPL to decltype and building with:
CC -std=c++11 -m64 -I/usr/local/include/boost-1_54 main.cc
But Solaris Studio 12.4's C++11 mode has other problems, so is there any way to somehow get a working __typeof__ or equivalent in Solaris Studio 12.4's C++03 mode?
(In case it makes any difference this is on Solaris 10 x86.)

I couldn't see that any of the items listed in the "Enforcement of C++ rules" section of the documentation would affect this program.
Here's a cut down version of the test program that doesn't use Boost:
#include <iostream>
#include <typeinfo>
// Helper to deduce the result type from a member function pointer.
template<typename T>
struct member_function_result_type
template<typename T, typename R, typename A1>
struct member_function_result_type<R (T::*)(A1) const>
    typedef R type;
template<typename T>
struct function_result_type
    typedef __typeof__(&T::operator()) F;
    typedef typename member_function_result_type<F>::type type;
template<typename R, typename A>
struct function_result_type<R (*)(A)>
    typedef R type;
struct my_functor
    float operator()(float numerator) const
        return numerator / 2.0f;
int main(int, char **)
    typedef function_result_type<my_functor>::type my_type;
    my_functor f;
    my_type ret = f(5.0f);
    if (typeid(my_type) == typeid(float))
        std::cout << "As expected " << ret << std::endl;
    return 0;
Interestingly this doesn't compile with either Solaris Studio 12.3 or Solaris Studio 12.4, but each gives different errors:
Solaris Studio 12.3:
CC -m64 main.cc
>> Assertion: unexpected type_builder::visit_unspec (../lnk/v2mangler.cc, line 1636)
    while processing main.cc at line 39.
Solaris Studio 12.4:
CC -m64 main.cc
"main.cc", line 20: Error: Unexpected type name "F" encountered.
1 Error(s) detected.
So Solaris Studio 12.4 is basically failing in the same way as the original test program that used the Boost wrappers. Solaris Studio 12.3 is failing with an assertion error that doesn't occur when the Boost wrappers are used.
It turns out that __typeof__ was only used in one place in our codebase (wrapped by Boost), so I've just removed it and implemented that code in a different way. So this isn't blocking upgrade of Solaris Studio, and we don't even have __typeof__ in our codebase any more, but I just thought you might be interested to have this program as a test case for future development.
It works fine with clang++ in C++03 mode on Mac OS X so I doubt the problem is caused by the source code not conforming to standard:
clang++ main.cc
./a.out
As expected 2.5
This case may be particularly tricky for a compiler to handle due to __typeof__ being applied to a pointer to a const member function.

12.4 Beta, Error: Could not find a match for std::_Tuple_impl 0, std::string && ::_Tuple_impl(std::tuple std::string &&

Hi,
would you guys say code that compiles fine without -std=c++11 should also compile *with* -std=c++11?
raider@sol112_x86:/tmp $ CC -V
CC: Sun C++ 5.13 SunOS_i386 Beta2 2014/06/17
raider@sol112_x86:/tmp $ CC buggy.cc
raider@sol112_x86:/tmp $ CC -std=c++11 buggy.cc
Error: Could not find a match for std::_Tuple_impl<0, std::string &&>::_Tuple_impl(std::tuple<std::string &&>, std::string ) needed in std::tuple<std::string &&>::tuple<std::string, void>(std::string &&).
"/opt/SolarisStudio12.4-beta_jul14-solaris-x86/lib/compilers/CC-gcc/include/c++/4.8.2/tuple", line 868:     Where: While instantiating "std::tuple<std::string &&>::tuple<std::string, void>(std::string &&)".
"/opt/SolarisStudio12.4-beta_jul14-solaris-x86/lib/compilers/CC-gcc/include/c++/4.8.2/tuple", line 868:     Where: Instantiated from std::forward_as_tuple<std::string>(std::string &&).
"/opt/SolarisStudio12.4-beta_jul14-solaris-x86/lib/compilers/CC-gcc/include/c++/4.8.2/bits/stl_map.h", line 485:     Where: Instantiated from non-template code.
1 Error(s) detected.
raider@sol112_x86:/tmp $ cat buggy.cc
#include <map>
#include <string>
typedef std::map<std::string, std::string> StrStrMap;
int main(void)
    StrStrMap dict;
    dict["bug"] = "feature";
    return 0;

C++11 is approximately a superset of C++03. If you write in the common subset, the code will compile in any mode. For example, a basic hello-world program
#include <iostream>
int main() { std::cout << "Hello, world!\n"; }
will compile and run as C++98/03, as C++11, as C++14, and I'm sure will also work with the next standard, planned for 2017.
But if you use syntax or library types and functions that are new in C++11, the code will not compile as C++03.
In my previous post, I might have sounded too negative about compiling C++03 code in C++11 mode. If you have a C++03 program that is intended to be portable, and that works with different compilers on different platforms, chances are good that it will continue to work in C++11. The chances are very good that only minor modifications will be needed.

Sun C++ 5.13 SunOS_sparc Beta2 2014/06/17: Error: Could not find a match for std::_Tuple_impl 0, unsigned&& ::_Tuple_impl(std::tuple unsigned&& , unsigned) needed in std::tuple unsigned&& ::tuple unsigned, void (unsigned&&).

I think I downloaded the latest Solaris Studio version. When I compile my code, I got the following error:
bash-3.00$ gmake -f Makefile.sun
CC -c -o metadata.o metadata.cpp -I. -I.. -g -O0 -std=c++11 -I/export/home/oracle/zhifan/boost_1_55_0 -I/export/home/oracle/zhifan/instantclient_11_2/sdk/include -V
CC: Sun C++ 5.13 SunOS_sparc Beta2 2014/06/17
ccfe: Sun C++ 5.13 SunOS_sparc Beta2 2014/06/17
"metadata.cpp", line 20: Warning: db hides databus::MetadataManager::db.
"metadata.cpp", line 26: Warning: db hides databus::MetadataManager::db.
Error: Could not find a match for std::_Tuple_impl<0, unsigned&&>::_Tuple_impl(std::tuple<unsigned&&>, unsigned) needed in std::tuple<unsigned&&>::tuple<unsigned, void>(unsigned&&).
"/export/home/oracle/zhifan/SolarisStudio12.4-beta_jul14-solaris-sparc/lib/compilers/CC-gcc/include/c++/4.8.2/tuple", line 868:     Where: While instantiating "std::tuple<unsigned&&>::tuple<unsigned, void>(unsigned&&)".
"/export/home/oracle/zhifan/SolarisStudio12.4-beta_jul14-solaris-sparc/lib/compilers/CC-gcc/include/c++/4.8.2/tuple", line 868:     Where: Instantiated from std::forward_as_tuple<unsigned>(unsigned&&).
"/export/home/oracle/zhifan/SolarisStudio12.4-beta_jul14-solaris-sparc/lib/compilers/CC-gcc/include/c++/4.8.2/bits/stl_map.h", line 485:     Where: Instantiated from non-template code.
1 Error(s) and 2 Warning(s) detected.
gmake: *** [metadata.o] Error 2
with the same code, It can be compiled with g++ on Linux platform.
1 Is the issue same as 12.4 Beta, Error: Could not find a match for std::_Tuple_impl<0, std::string &&>::_Tuple_impl(std::tuple<std::string &&> ?
2 Do we have any workaround for this issue?
3 When will the final 12.4 release be available?

I can't say for sure whether your problem is fixed in the final release without a test case, but I'd say the chances are very good.
The final release will be announced soon. Watch this space, or the Oracle Solaris Studio page for an announcement.

Error C1189: #error : ERROR: You must use a GNU Compiler.

Hi everyone!
I have just ran into an error in which I don't know how to resolve. I'm trying to compile some code I have but it gives me the following : error C1189: #error : ERROR: You must use a GNU Compiler.
Before receiving this error I had received one stating that the header file "_mingw.h" was missing. I located that file and placed it in the correct folder. I also updated the project properties to locate the header file. The error itself says
that I must use a GNU Compiler but I'm not exactly sure how to do this.I do have mingw installed on the PC though
The error points to this part of the code :
#define __MINGW32_VERSION           3.2
#define __MINGW32_MAJOR_VERSION     3
#define __MINGW32_MINOR_VERSION     20
#define __MINGW32_PATCHLEVEL        0
#ifndef __GNUC__
#error ERROR: You must use a GNU Compiler.
#endif
#if (__GNUC__ < 3 || !defined(__GNUC_MINOR__) || (__GNUC__ == 3 && __GNUC_MINOR__ < 4) || (__GNUC__ == 3 && __GNUC_MINOR__ == 4 && __GNUC_PATCHLEVEL__ < 5))
#error ERROR: You must use a GNU Compiler version >= 3.4.5.
#endif
Also to note, I have the most recent version 4.8.4.1 of the compiler.
Any ideas on what I should so to fix this? Did placing the _mingw.h file in the folder have anything to do with it?
I would appreciate any help!! :-)

It appears that the author of the code you are compiling intended it to be compiled with the GNU compiler, not the Visual C++ compiler. So they manually inserted an error message (guarded by the #ifndef __GNUC__ check) so that if you compiled it on
Visual Studio, you'd be alerted to that fact.
I'm afraid you are now faced with either porting this code or finding a Windows distribution of it that is compatible with Visual Studio.
*NOT RECOMMENDED*, but to begin the process of porting it , simply delete the #error line from the source code and recompile with visual studio. You can expect millions of specific issues will need further attention. After all, it's probably
there for a reason. If you are new to this, that's probably NOT what you want to do.
On the other hand, what were you trying to compile that needed _mingw.h? That's the piece that needs to be configured correctly. You are likely not building it as the author intended.

Error while Compiling Form using f60gen

Hello All,
I am trying to compile a customized form in apps middle tier using f60gen command.
Actually the form is a customized version of FNDCPMCP.fmb(new form is in another name).
f60gen throws the following error,
Compiling package body CONC_PROG_PKG...
Compilation error on package body CONC_PROG_PKG:
PL/SQL ERROR 306 at line 4, column 3
wrong number or types of arguments in call to 'INSERT_ROW'
PL/SQL ERROR 0 at line 4, column 3
Statement ignored
PL/SQL ERROR 306 at line 59, column 7
wrong number or types of arguments in call to 'LOCK_ROW'
PL/SQL ERROR 0 at line 59, column 7
Statement ignored
PL/SQL ERROR 306 at line 109, column 3
wrong number or types of arguments in call to 'UPDATE_ROW'
PL/SQL ERROR 0 at line 109, column 3
Statement ignored
The form calls FND_CONCURRENT_PROGRAMS_PKG.
What may be the issue.
Thanks&Regards
Kannan Balasubramanian

Hello,
It seems that you are trying to call Functions or Procedures with an incorrect argument list.
Check the Function/Procedeure declaration.
Francois

CompileException: Error compiling web service (using ant jwsc)

Hi,
I try to use the ant task jwsc to create my portable artefacts for a web service.
Unfortunately I get the following exception:
BUILD FAILED
C:\Sun\jwsdp-2.0\apache-ant\bin\build.xml:7: weblogic.wsee.tools.jws.build.CompileException: Error compiling web service: C:\Sun\jwsdp-2.0\apache-ant\bin\src\test\WsTest.java
I do not know why the ant task can not compile my jws file (shown below) whereas Eclipse can!
package test;
import javax.jws.WebService;
@WebService (name="WsTestName", serviceName="WsTestServiceName")
public class WsTest {
* @param name
* @return Say hello to the person.
public String sayHello(String name) {
return "Hello, " + name + "!";
I hope somebody can help me.
Thanks
Franz

I thought I would post an update since I found a solution. The problem was a ClassCast exception in a part of the SSL stack, that wanted to use the ant BuildException class. This has the effect of hiding the real issue which was that the SSL connection was not successfully occuring. The real problem was that the SSL connection started with a WSDL retrieved via SSL, but the connection for the port was through a username/password. When a username/password is used to create a port, the WL stack falls back to http and causes a ClassCast exception on weblogic.wsee.connection.transport.http.HttpTransportInfo. The solution is to create an https transport object when the service impl is created:
HttpsTransportInfo transport = new HttpsTransportInfo ();
transport.setUsername (user.getBytes ());
transport.setPassword (pass.getBytes ());
gServiceImpl = new PersistenceManagerService_Impl (url, transport);
and to create the port without parameters:
port = getServiceImpl ().getPersistenceManagerServicePort ();
This allows one-way SSL with username/password for the connection.

Error compiling movie 'unknown error' premiere elements 11, tried every troubleshoot, what now?

I'm trying to export a video and I keep on running into this error message, which stops the export process. I've already visited and tried the troubleshooting suggestions on this page and this page, and this page, and none of those suggestions have worked. Here's a list of what failed:
Export in various file formats, every single one failed
I removed each asset individually and tried to export the video after each removal (100+ assets, imagine how long that took) and I got the error after every single time
I removed the front half of the video and tried exporting, still failed.
I removed the back half of the video (because if it failed if I removed the front, the problem asset should be in the back, right? nope.) and tried exporting, it still failed.
I don't have any Auto Color, Auto Contrast, Auto Levels, or Shadow/Highlight on any of my clips, although if I did, I'd try removing them.
All of my video files are the same format, so it's not a mix of formats in my timeline which is causing the problem.
I tried disabling auto-save, but it already tried to auto-save during the first 20 times I've attempted this, so the damage might already be done.
I obviously didn't use any other programs during this process
I moved all my assets to a different location on my harddrive, then reopened the project and refound the assests one by one, each of which had no problem getting into the timeline. I tried rendering after this, and the render bar sailed along until a certain point, where it got stuck indefinitely. I cancelled the render, which froze the program. I had to close it and re-open the project, and re -link every asset again. Once again there were no problems adding the assets. I saved the project before the next render, so I wouldn't have to do this a third time, and it was a good idea I did, because the program froze again during the subsequent render.
My computer didn't go to sleep or hibernate during the first 5-10 times I tried various methods for exporting. It has slept in some of the 10+ times I've tried it since, because I don't have the patience to sit in front of my screen and watch it fail again and again and again and again and again and again and again and again and again and again and again and again and again and again and again and again and again and again and again. CAN YOU TELL THAT I'M FRUSTRATED???
The first thing I noticed that was wonky with this project was when I first tried to render my sequence. The render window chugged along just fine until it got to a certain point (can't remember the %), at which point it got stuck indefinitely. I had to cancel the render, and then render other parts of the sequence bit by bit, avoiding what I thought was the problem part. I tried rendering the problem part, and it got stuck again, so I eventually just decided that that clip didn't need to be in this movie, so I deleted it from the timeline. All should be good at this point, my whole timeline is green from being rendered. But, when I tried to export the video, I started down this tourtuous path of failed exports and frustration, and now here I am, posting in the forum, hoping someone can help me out.
The problem asset (if that was the problem in the first place) isn't in the timeline any more, but I'm still having issues exporting the movie. The fact that rendering still gets stuck after re-linking assets, and after getting rid of the first problem asset, makes me think that there is some bigger underlying issue, but I have no idea what that could be. I followed these instructions and re-opened the project, tried to export, and still got that 'Unknown Error'.
Please assist before I put my fist through this company computer.

maxlong99
Thanks for your update. Let us go through it.
You wrote
During one of my isolation tests, I stripped the timeline off all assets except clips from the same footage and I also removed that New Blue effect, and the project exported without issue.
Did you run a control which included "with the New Blue effect" in this isolated test which does not sound like the "Full Project" as suggested in what you wrote next?
You wrote
Could multiple renders and project files be part of the problem? I've created many, many different versions of this project to test and see where the problem is, which in turn means I've created many different render files for each different project. If this could be part of the problem, should I keep the original project and delete all my test projects and associated render files therein?
Each of the projects should have its own name which will be applied to its files. And each project should have rendered files specific to it (for SD project whether video or still clip = DV .avi file; for HD project whether video or still clip = MPEG2.mpg file). Are you naming these different projects with distinctive names?
Have you expanded the Timeline with the -+ slider and closely inspected the track content for
a. leftovers such a very thin vertical black lines scattered about
b. gaps
c. transitions, and especially audio placement
You have probably seen the following, but I will post the link just in case. I suspect you have been there and done that.
http://helpx.adobe.com/premiere-elements/kb/error-error-compiling-movie-render.html
Please let us know what the closer inspection of the Timeline content (as mentioned) revealed.
Thanks.
ATR

"error compiling movie: unknown error" in premiere cs3.2 / win8

This happens during both Export and Render Work Area, as soon as the renderer hits one of my assets. Initially, this asset was an uncompressed AVI with an alpha channel, over another uncompressed AVI. You can see the Progress bar suddenly get extremely slow when it hits that point, and then it gives the unknown error within about 10 frames. As far as I remember, this NEVER happened when I was rendering on my laptop, running Windows 7 (a former temporary setup while I was waiting to get a new workstation).
I've tried taking some pressure off Premiere... I rendered the sequence in AfterEffects, so Premiere didn't have to deal with the compositing. I also wrapped the asset in a MOV container, with PNG compression, so the files weren't so large. This seemed to help in certain cases -- using this method, I can get through some of the biggest assets. It still happens a bunch of times throughout my sequence, though.
I have the rendering optimized for "memory" in the project settings. I also have more than enough hard drive space on the drive I'm exporting to.
My instinct and limited powers of observation suggest to me that this is a memory problem. This computer's not exactly a world-class production system. However, there are other clips -- mostly HD footage directly from the camera (HVX200) but also some shorter-duration uncompressed AVI's -- that seem to render fine. Plus, even some clips already recompressed to MOV's in AE are causing the same problem. I don't want to have to fiddle with this on every project til I can afford Adobe CC... has anyone else seen problems like this, and found some kind of broader solution to them?

If you have not seen this Adobe KB article on Error Compiling Movie, it might also be helpful: http://helpx.adobe.com/premiere-pro/kb/error-compiling-movie-rendering-or.html
Good luck,
Hunt

Error Compiling Movie: Unknown Error - Simple Solution in Premiere

ADOBE PREMIERE EDITORS: Be forewarned about the infamous "Error Compiling Movie: Unknown Error." It usually doesn't happen unless you're editing a big project like I am and then it screws with you because you've been editing fine until it's time to export and you can't even make your project into a movie, which defeats the purpose of all your hard work. I looked up forum after forum and finally there were a-few solutions that worked for some and not for others. (Adobe said they will get their team on this case right away.) Let's just say I do not have Final Cut Pro, so I cannot export to XML and then import into FCP to do the exporting like some people found helpful. Nor do I have the time to import my Premiere project into After Effects only to manually create all of my "dip to white" transitions. But I would have had to do that if nothing else I tried worked and thankfully, I was able to trick Premiere. It's so simple, it's stupid. Just open a new project in Premiere. Then "import" the project you wanted to export. For some odd reason, without changing anything of substance or in other words, doing practically nothing about the issue, it exports now. Just because I opened a "new" project and "imported" my old project. Of course, this may not work for others, but please try it and then reply your results. Me and I'm sure Adobe would find it very helpful to see if that worked for most people. It worked for me.
[Profanity is not permitted. Removed. Please keep your language family-friendly.]

I agree, Jim, but my train of thought was that a lot of people were changing and replacing and taking out effects and clips from their sequence to see if it was the components in the sequence that was causing the "unknown error." However, I had the same "unknown error" and was able to export in Premiere anyway by just practically "telling" Premiere that it's a new project, when I'm importing an old project to do the same task. Which makes me think that maybe Premiere doesn't have a problem exporting the specific files that most people thought were the problem, maybe it was just a glitch somewhere in Premiere that made the software think there was a problem after a certain amount of time, when there really wasn't. I don't know, I'm not a coder so that's just my layman guess. I did notice that Premiere was exporting/rendering fine when I was just beginning my project, but after it got to a certain length, that's when I couldn't render or export anymore. It honestly baffles me too. It would be useful to see what the results are if other people tried this simple switch-a-roo so possible issues can be narrowed down.

Error compiling movie: out of memory !!! Premiere CS3

Hey every1 --- desperate need of some guidance here. I've been all over the internet and these boards --- have seen this come up with a lot of people but no real solid answers that I'd seen or that have helped.
First off, my comp is an imac 3.06 Ghz Core 2 Duo, 4GB memory. (just bought the computer last year)
I'm shooting with a canon 7d in HD (1920 x 1080p) and I'm editing using Premiere CS3. I've noticed that editing has become very difficult since I started using HD; playback is very choppy and the 'quality settings' in the preview monitor are a joke and do absolutely nothing whether it's on draft or automatic.
Anyway, despite the choppiness and slow editing, I was able to get a 3 min clip completed. These are my export settings:
Export > Movie
H.264
960 x 540
23.976 fps
Square pixels,
100% quality,
recompress: unchecked.
Once exported (I was able to export smaller timelines previously--- say, 50 secs of footage or so) the files look AMAZING and from there I use them for online video via vimeo -- stuff looks really awesome.
Anyway, recently I'm getting: 'Error compiling movie. Out of Memory. Change optimization preferences to memory', etc. etc. Well, I did as it asked and that did absolutely nothing.
So, I decided to open up my activity monitor and watch what's going on with my memory while premiere is exporting. I also closed ALL unused applications. System runs at about 2.16gb free memory before I start exporting. Memory seems stable during export UNTIL around the 50% exporting mark, and then its starts climbing like crazy and my memory just gets eaten up --- finally resulting in a crash at about 70% render.
I really don't know what else to do. Does my computer not have the power to export 3 min of HD footage??? This seems surprising to me ... I don't know what else to try.
Thanks for any help !!!
-Mark

Hi Mark,
Hope you are having a wonderful day.
Sorry to hear that you are having such a hard time. Although I did not specifically get the out of memory error, I can only be convinced that that is the problem.
Although I am not a computer whiz, I have learned my share of things. I have been troubleshooting a video all this week (really, a couple of months), but I had to put it aside and come back to it.
I tend to be talkative; so I will get back on topic or at least try to stay on topic, so bare with me. Smile....
Here is what I have learned and maybe it will help you with yours if you have not found a solution already.
Based on your hard drive specs and ram, I was thinking the same thing e.g. hard to believe that you are not able to render 3 minutes of video in HD. I primarily work in HD/1080p.
The place where we are similar is that I eventually could get the video to render by reducing the workspace that is rendered piece by piece until I was able to render the entire file. When I tried to encode the video, I kept getting error compiling messages. I tried different codecs, deleting xmpses, not rendering xmpses, etc. I could not figure out the keyframe trick so I just put in a black video from after effects. By the way, all of my videos intro the same and no problems with the other videos fading from black to a still photo.
Here is the problem from what I could gather.
Simply, out of memory. I was rendering and trying to encode a 45 minute video with after effects compositions and the dynamic linking kept timing out once the file reached a certain size. There were no other true fancy effects. I do alot on the green screen and just use after effects to make a white background. Smaller videos, rendering has been slow, but doable. This massive video would never compile. I won't bore you with my system specs, but know that similar to your situation, the system in theory should have been adequate.
Work around...
I ended up exporting the video as 3 separate 15 minute video clips. I could not get them to encode as .wmv files so I encoded them as .mp4 files. By the way, I still encoded in HD.
Here is the kool stuff.
After what seems like forever, I found a converter that could convert .mp4 to .wmv and rejoin the files to play as one large file with the same quality as my source file and no audio synching issues. By the way, I do not work for this company. I just found this encoder at 4 o'clock this morning after trying all of these other converters that were straight garbage or were free to try and not really free if you wanted quality. By the way, the conversion tool is not only user friendly, but free which is awesome.
The name of the software is Media Cope. Just visit www.mediacope.com.... Be sure to send the owner an e-mail to thank him and let him know that Mark sent you. I just sent him an e-mail earlier to let him know that I was going to post on Adobe what a wonderful software he created. I just needed something free, quick, easy, and user-friendly.
Whoops... got off topic....
I'll do a summary at the end for my recommendation to you.
Here are some other tidbits that I learned as well that may save you some time.
When editing this is what I do now...
1) I use CCleaner from www.piriform.com to quickly clean my system especially temporary files before I start editing and if I am switching from project to project.
2) In all of my troubleshooting this week, I found a kool little software call Smart Close. (Just google to find this one.) It closes all of your non-critical background files in order to maximize system resources. (When I am done editing, I just reboot and all of my background programs start up again. You do not have to do it that way, that's just how I do it. The nice thing is that it is a wizard interface; so you do not have to try to figure out if a process in the background is critical or non-critical.)
3) Depending on my mood and if I did not forget, smile.... I change the priority to high for the adobe products in the task manager.
More things that I learned...
1) We are not able to change the temp file for how Adobe encodes. My logic was if I could change the encoding to a hard drive that was simply empty so Adobe can render there, then I would be good. Wrong.... Adobe does not allow you to change the temp location. (Note: This is what I discovered in the different forums; so encoding to an external hard drive is not the best move. Read more to se why.)
2) It is best to have your scratch disks as your internal disk. If you have multiple internal disks, then you shoud check some other forums for the best way to set up your scratch disks. When I was just starting out, I bought an external hard drive enclosure with the thought that I would work primarily off of those external drives with the option of doing some RAID work. Bad idea... Even the simplest of tasks took forever. It was best to work off of the internal drive. By the way, I still use the enclosure hard drives for backups in order to allow me to free up some space on the primary hard drive. When rendering, it is truely a use of the RAM and the hard drive space in the editing process.
I'll leave you with this...
I take back what I said earlier, you do need a new computer with more RAM. If you can get 16 GB, great. The primary reason is that if you are editing in Premiere and if you do anything with dynamic linking into your after effects, it will cause your ram to climb as the file gets larger. Of course, be sure to get the largest hard drive possible as an internal drive, the maximum on processor speed, 64 bit operating system, etc.
I know you are like by now, Mark, what is the point of all of your gibberish... Really, it was to share with you and others on some really kool work arounds.
My recommendation for you are as follows.
1) You have already proven that you can render and encode 50 second clips and you are very happy with those clips, so create separate smaller clips for this particular project.
2) Once you have the final clips outputted, then use Media Cope to rejoin the files as one large file and you are done.
(Note: The one thing that I did notice is that there is a small break right between where the files are merged. The nice thing is that as a video editor/film maker, you will be the only person likely to notice it because you know where the breaks are. The other nice thing is that the untrained eye will not notice it and if you had to submit your work to a television station especially on DVD, the video system should recognize the file as one large file and play all the way through. Note: For the last sentence, I said 'should play all the way through' as I have not had a chance to check this one yet as I just finished this project today and still have to turn in all of my videos. I still have some more videos to merge.)
By the way, the only specs that I will share are that I work on a PC and it has 8 GB of ram.
The summary of the summary is that sometimes an error compiling message could simply be that your system does not have resources to render and the encode that large of a file. At least that is what I learned this week or at least my specific situation. The confirmation is that I just finished shooting a documentary which was one hour and some change and was able to render and encode to .wmv with no problems. Of course, I was not using after effects for this particular project.
In either case, hope all the information helps you with your project.
Have a wonderful day and happy holidays everyone.
Mark

Compiling / Linking error in Solaris Studio 12.3 C++ Compiler

Hi,
I downloaded Solaris Studio 12.3 (SolarisStudio12.3-solaris-x86-pkg.tar.bz2) & unpacked and installed on the default directory (in non-interactive mode) of local machine running Solaris 10 10/09 s10x_u8wos_08a X86. Also applied all the required patches by running the "install_patches.sh" after running "solarisstudio.sh".
My application has make files to take care of the building the whole application. I made necessary changes to the make files to pick up the latest Solaris 12.3 C++ compiler. I could see that it has picked up for compiling.
While building the whole project, the source files are compiled to object files successfully but when it is creating shared object (.so) files it stops at the below error:
/opt/solarisstudio12.3/prod/bin/CC -mt -i -library=stlport4 -zdefs -ztext -G -h xmem.so -Lpic/lib/debug/sol32 -L/MyHome/open_source/abc/lib/sol32 -L/MyHome/open_source/mno/lib/sol32 -L/MyHome/open_source/pqr/lib/sol32 -L/MyHome/open_source/xyz/lib/sol32 -g -lc -lCrun -lumem -m32 -o debug/xmem.so
*usage: CC [ options ] files. Use 'CC -flags' for details*
Note:
Some third party open source libraries are being used in the project. Those libraries are not built using Solaris Studio 12.3 compilers.
Tried changing lot of options and re-ordering the switches/options but no luck. But on changing as mentioned below the shared library DID build successfully
- Added space after '-L' option, added '*' at the end of path like "-L/MyHome/open_source/pqr/lib/sol32/*".
Note:*_
+1) I tried compiling a separate small program into object files using Solaris 12.3 C++ compiler but got the above error/problem when linking the experimental libraries (shared object) with most of the options used in the above command.+
+2) The same project gets built successfully in Solaris Studio 12.1 (studio12.1-091123) and Solaris Studio 12 (studio12-070724) but the same is failing in Solaris Studio 12.3.+
Can someone please help me to understand the issue and resolve it? Let me know if any more information is needed.
Thanks in advance.
-Vijay
Edited by: 974820 on Dec 4, 2012 7:05 PM

Hi Steve,
Thanks for the valuable input on the -xdebugformat option. Every suggestion of yours is helping me to proceed further to some extent. This time too, after removing the "-xdebugformat=stabs", I am able to go past the previous error.
BUT this time I am running into -xarch option problem. I am not sure if this is a application design or compilation options issue or compiler issue. Posting it here to get some inputs to understand more on this before raising any service request with Oracle Support.
The error details and workarounds used with their result._
NOTE: All these compilation were passed in earlier Studio versions.
ORIGINAL - with -xarch=sse2
===============================ERRORS===============================
/opt/solarisstudio12.3/prod/bin/CC -errtags=yes -Kpic -mt -errwarn -xdepend=yes -DUSING_STUDIO11 common/xm2/x86/common/inc/templates.il -D__EXTENSIONS__ -g -xarch=sse2 -Icommon/xccolor/inc -Icommon/xm2/x86/include -Icommon/xm2/x86/assembly/inc -Icommon/xm2/x86/common/inc -Iinclude -Icommon/xm2/x86hbc/expander/inc -Icommon/xm2/x86/jpeg/inc -Icommon/xm2embedded/host/x86/hbc/inc -Icommon/xm2/common/inc -Iimg-drv//include -Icommon/xm2/config/inc -Icommon/Utilities/include -Iimg-drv/ipshared/include -Icommon/cpp/inc -I/workspaces/3rd_Party/open_source/poco-1.3.6p2/include -I/workspaces/3rd_Party/open_source/boost_1_49_0/include -I/workspaces/3rd_Party/open_source/ACE-6.0 -I/workspaces/osnl/include -I/workspaces/osnl/SunOS/include -m32 -c common/xm2/x86/hbc/src/hbc_jpeg.c -o artifacts/common/xm2/x86/obj/debug/sol32/hbc_jpeg.o
Assembler:
"/tmp/yabeAAAIaaOxX", line 294 : Syntax error
Near line: " maskmovq %mm0,%ebp"
"/tmp/yabeAAAIaaOxX", line 1175 : Syntax error
Near line: " maskmovq %mm0,%ebp"
Failure in /opt/solarisstudio12.3/prod/bin/fbe, status = 0x7f00
Fatal Error exec'ing /opt/solarisstudio12.3/prod/bin/fbe
common/xm2embedded/host/x86/src/hbc_jpeg.c:1062 Warning: mmextract_epi16 intrinsic requires -xarch=sse4_1.*
cc: acomp failed for common/xm2/x86/hbc/src/hbc_jpeg.c
gmake[1]: *** [_artifacts_/common/xm2/x86/obj/debug/sol32/hbc_jpeg.o] Error 2
gmake[1]: *** Waiting for unfinished jobs....
===============================ERRORS===============================
AFTER CHANGING -xarch=sse2 to -xarch=sse4_1
===============================ERRORS===============================
/opt/solarisstudio12.3/prod/bin/CC -errtags=yes -Kpic -mt -errwarn -library=stlport4 -DCPU_LITTLE_ENDIAN -DDSP_LITTLE_ENDIAN -xarch=sse4_1 -DTIXML_USE_STL=1 -erroff=doubunder -g -Icommon/xm2/xm2util/src -Icommon/xm2/xm2util/inc -Icommon/xm2/common/inc -Icommon/xm2/config/inc -Icommon/xm2embedded/host/utils/inc -Icommon/xm2/imageutil/inc -Icommon/xm2/x86hbc/include -Icommon/include -Iinclude -Icommon/xcc/inc -Icommon/cpp/inc -Icommon/Util/include -I/workspaces/open_source/poco-1.3.6p2/include -I/workspaces/3rd_Party/open_source/boost_1_49_0/include -I/workspaces/3rd_Party/open_source/ACE-6.0 -I/workspaces/CP.00_compiler/docusp/osnlayer/include -I/workspaces/osnl/SunOS/include -m32 -c common/xm2/xm2/src/JPEGHelpers.cpp -o artifacts/common/xm2/xm2/src/obj/debug/sol32/JPEGHelpers.o
"common/xm2/xm2/src/JPEGHelpers.cpp", line 918: Error, nostructsym: Variable table is not a structure.
"common/xm2/xm2/src/JPEGHelpers.cpp", line 919: Error, nostructsym: Variable table is not a structure.
"common/xm2/xm2/src/JPEGHelpers.cpp", line 926: Error, nostructsym: Variable clear_high2bits is not a structure.
"common/xm2/xm2/src/JPEGHelpers.cpp", line 927: Error, nostructsym: Variable clear_high2bits is not a structure.
"common/xm2/xm2/src/JPEGHelpers.cpp", line 1223: Error, unassigned: The variable clear_high2bits has not yet been assigned a value.
"common/xm2/xm2/src/JPEGHelpers.cpp", line 1224: Error, unassigned: The variable table has not yet been assigned a value.
6 Error(s) detected.
/opt/solarisstudio12.3/prod/bin/CC -errtags=yes -Kpic -mt -errwarn -library=stlport4 -DCPU_LITTLE_ENDIAN -DDSP_LITTLE_ENDIAN -xarch=sse4_1 -DTIXML_USE_STL=1 -erroff=doubunder -g -Icommon/xm2/xm2/src -Icommon/xm2/xm2/inc -Icommon/xm2/common/inc -Icommon/xm2/config/inc -Icommon/xm2/utils/inc -Icommon/xm2/imageutil/inc -Icommon/xm2/x86/include -Icommon/xm2/include -Iinclude -Icommon/xcc/inc -Icommon/cpp/inc -Icommon/Util/include -I/workspaces/3rd_Party/open_source/poco-1.3.6p2/include -I/workspaces/3rd_Party/open_source/boost_1_49_0/include -I/workspaces/3rd_Party/open_source/ACE-6.0 -I/workspaces/osnl/include -I/workspaces/osnl/SunOS/include -m32 -c common/xm2/xm2/src/Color.cpp -o artifacts/common/xm2/xm2/src/obj/debug/sol32/Color.o
"common/xm2/xm2/inc/JPEGStripsImage.hpp", line 85: Error, badinitlval: Initializing std::vector<unsigned, xm2::XM2Allocator<unsigned>>& requires an lvalue.
"common/xm2/image/inc/ISelector.hpp", line 36: Error, badinitlval: Initializing std::vector<unsigned, xm2::XM2Allocator<unsigned>>& requires an lvalue.
"common/xm2/image/inc/RasterSelectorImage.hpp", line 53: Error, badinitlval: Initializing std::vector<unsigned, xm2::XM2Allocator<unsigned>>& requires an lvalue.
3 Error(s) detected.
6 Error(s) detected.gmake[1]: *** [_artifacts_/common/xm2/xm2/src/obj/debug/sol32/JPEGHelpers.o] Error 2
gmake[1]: *** Waiting for unfinished jobs....
3 Error(s) detected.gmake[1]: *** [_artifacts_/common/xm2/xm2/src/obj/debug/sol32/Color.o] Error 2
===============================ERRORS===============================
Thank you.
Vijay

"Error compiling movie. Unknown error" - my solution

I was trying to encode a project to mpeg-2 DVD yesterday, when I repeatedly received the message "Error compiling movie. Unknown error." I had had this happen before and I can't remember what hoops I jumped through to fix it. But yesterday, I found a simple solution that may work for someone else in the future should they ever search for this problem in our forum.
The problem turned out to be with some .wmv files mixed in with all the other .avi files I was using. And the solution was to right-click on the .wmv clips in the project bin, choose "modify," then choose "interpret footage. From there, I changed the frame rate choice from "Use frame rate from file" (which was 30.00) to "Assume this frame rate" and entered a value of 29.97. Voila! Everything rendered without a hitch!
Hope this helps someone.
J. D.

I agree, Jim, but my train of thought was that a lot of people were changing and replacing and taking out effects and clips from their sequence to see if it was the components in the sequence that was causing the "unknown error." However, I had the same "unknown error" and was able to export in Premiere anyway by just practically "telling" Premiere that it's a new project, when I'm importing an old project to do the same task. Which makes me think that maybe Premiere doesn't have a problem exporting the specific files that most people thought were the problem, maybe it was just a glitch somewhere in Premiere that made the software think there was a problem after a certain amount of time, when there really wasn't. I don't know, I'm not a coder so that's just my layman guess. I did notice that Premiere was exporting/rendering fine when I was just beginning my project, but after it got to a certain length, that's when I couldn't render or export anymore. It honestly baffles me too. It would be useful to see what the results are if other people tried this simple switch-a-roo so possible issues can be narrowed down.

Error compiling certain uses of std::tuple in Solaris Studio 12.4 beta July refresh

Similar Messages

Maybe you are looking for