bug-commoncpp
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Successfully compiled 1.3.1 on AIX 5.2 /w Visual Age C++ 6.0 after s


From: David Sugar
Subject: Re: Successfully compiled 1.3.1 on AIX 5.2 /w Visual Age C++ 6.0 after some tweaking
Date: Fri, 28 Jan 2005 10:21:30 -0500
User-agent: Mozilla Thunderbird 1.0 (Macintosh/20041206)

Okay that is odd. I thought the atomic stuff we did was correct for aix. Fortunately we have not started work yet on framework classes for lockfree data structures in Common C++ (those require atomics...). Lockfree data structures is something that was proposed which I do wish to look at in the near future, though.

In any case I will wait to hear what progress you make.

David

address@hidden wrote:
Here is two more items:

Believe it or not but I needed to comment out HAVE_ATOMIC_AIX.  It was causing core 
dumps.   I also commented out HAVE_ALLOCA_H to get rid of warning messages about 
"pragma __alloca not defined" go figure?

Bad news: my app just core dumped on an old favorite "if (th->priv->_type == 
threadTypeDummy)":

Segmentation fault in ThreadDestructor(ost::Thread*) at line 1125 in file "" 
($t373)
could not read "thread.cpp"
(dbx) where
ThreadDestructor(ost::Thread*)(th = 0x3058d428), line 1125 in "thread.cpp"
ccxx_thread_destructor(void*)(arg = 0x3058d428), line 1131 in "thread.cpp"
pth_key._specific_data_cleanup(??) at 0xd0067420
pth_pthread.pthread_exit(??) at 0xd0058030
ThreadExecHandler(ost::Thread*)(th = 0x3058d428), line 1115 in "thread.cpp"
ccxx_exec_handler(ost::Thread*)(th = 0x3058d428), line 1136 in "thread.cpp"
pth_pthread._pthread_body(??) at 0xd0052588

I have a few ideas on how to fix this, let me experiment a little.

Ron

-----Original Message-----
From: David Sugar [mailto:address@hidden
Sent: Thursday, January 27, 2005 3:13 PM
To: Horrell, Ronald (Card Services)
Cc: address@hidden
Subject: Re: Successfully compiled 1.3.1 on AIX 5.2 /w Visual Age C++
6.0 after some tweaking


True...looks like the only other place its needed is in string.h.  Done.

address@hidden wrote:

In misc.h the pragma pack is wrapped with "#if defined(__GNUC__) || 
!defined(__hpux)" but it isn't in cc++/string.h.  The easiest fix would be to wrap 
it also.

Ron

-----Original Message-----
From: David Sugar [mailto:address@hidden
Sent: Thursday, January 27, 2005 2:54 PM
To: Horrell, Ronald (Card Services)
Cc: address@hidden
Subject: Re: Successfully compiled 1.3.1 on AIX 5.2 /w Visual Age C++
6.0 after some tweaking


I should add we need a reliablle define to test for VA compiler. I believe we currently use an _AIX macro to detect and select some aix specific things, but I would not want to apply that macro check alone as > some of these changes would break using gcc on aix. The behavior of pragma pack(unpack), for example, is specific to the VA compiler.

address@hidden wrote:


Hi All,
        I was finally able to successfully compile version 1.3.1 on AIX 5.2 /w 
Visual Age C++ after some tweaking.

System Details:

 Compiler: VisualAge C++ Compiler Version 6.0.0.5
 OS: AIX 5.2.0.0

C/C++ Flags: "-qcpluscmt" or "-+" Allows C++ comments "//" in C files. "-qlanglvl=redefmac" Allows macros to be redefined without #undef "_THREAD_SAFE"

configure Flags:
 "--enable-static --disable-shared": Personal preference
 "--without-exceptions": Needed due to use of "dynamic_cast without RTTI" error.

config.h:
 HAVE_SYS_LIBCSYS_H: Had to be commented out due to errors with memcopy 
declarations.

libtool:
 "${CC/${host}-}": "bad substitution error" (lines 904 and 1297) had to comment 
out check.


With these changes it compiles and runs successfully.

Note:
 The VisualAge C++ Compiler doesn't like "#pragma pack()" and gives numerous warning 
messages. With this compiler you would need to do "#pragma pack(nopack)" instead.

Thanks,
Ron Horrell


**********************************************************************
This transmission may contain information that is privileged, confidential 
and/or exempt from disclosure under applicable law. If you are not the intended 
recipient, you are hereby notified that any disclosure, copying, distribution, 
or use of the information contained herein (including any reliance thereon) is 
STRICTLY PROHIBITED. If you received this transmission in error, please 
immediately contact the sender and destroy the material in its entirety, 
whether in electronic or hard copy format. Thank you
**********************************************************************



_______________________________________________
Bug-commoncpp mailing list
address@hidden
http://lists.gnu.org/mailman/listinfo/bug-commoncpp





**********************************************************************
This transmission may contain information that is privileged, confidential 
and/or exempt from disclosure under applicable law. If you are not the intended 
recipient, you are hereby notified that any disclosure, copying, distribution, 
or use of the information contained herein (including any reliance thereon) is 
STRICTLY PROHIBITED. If you received this transmission in error, please 
immediately contact the sender and destroy the material in its entirety, 
whether in electronic or hard copy format. Thank you
**********************************************************************






reply via email to

[Prev in Thread] Current Thread [Next in Thread]