Sun CC weird compiling message : is out-of-date and will be regenerated

[modemer]

Hello,

I got weird compiling message similar like following when I compiled my
simple code on Sun 5.8 with CC WorkShop 6 update 2 C++ 5.3.

CC -g -o myclass.o -c myclass.cpp
CC -g -o main.o -c main.cpp
CC -g -L/usr/lib -lsocket -lnsl -o main myclass.o main.o
__1cH__rwstdJ__rb_tree4CpknJmyclassHCSOCKET_nDstdEpair4Ck4n0BNCSOCKEVENTMGRJ
SOCKEVENT___n0AL__select1st4n0G_C4__n0DEless4C4__n0DJallocator4n0G____Efind6
kMr6_n0KIiterator__ is out-of-date and will be regenerated

These messages didn't stop the compiling, the binary is generated correctly.
But I wonder why these messages displayed?

One thing I have to mention is that, in the code, I use STL map template, I
guess STL could be a suspect. I am thinking maybe my compiling line should
do something for STL?

Thanks in advance for your help.

M.

 

[Victor Bazarov]

I got weird compiling message similar like following when I compiled my
simple code on Sun 5.8 with CC WorkShop 6 update 2 C++ 5.3.
[..]
These messages didn't stop the compiling, the binary is generated correctly.
But I wonder why these messages displayed?
[..]

Messages are compiler-specific and are not defined by the language itself.
You might find it more useful to ask in a newsgroup that deals with your
compiler, where compiler-specific inquiries are on topic.

V

 

[Panjandrum]

Hello,

I got weird compiling message similar like following when I compiled my
simple code on Sun 5.8 with CC WorkShop 6 update 2 C++ 5.3.

CC -g -o myclass.o -c myclass.cpp
CC -g -o main.o -c main.cpp
CC -g -L/usr/lib -lsocket -lnsl -o main myclass.o main.o
__1cH__rwstdJ__rb_tree4CpknJmyclassHCSOCKET_nDstdEpair4Ck4n0BNCSOCKEVENTMGRJ
SOCKEVENT___n0AL__select1st4n0G_C4__n0DEless4C4__n0DJallocator4n0G____Efind6
kMr6_n0KIiterator__ is out-of-date and will be regenerated

These messages didn't stop the compiling, the binary is generated correctly.
But I wonder why these messages displayed?

Maybe generated template code is stored in separate files, which are
re-generated when the compiler deems so?? Look around, do you see
suspicious files?

One thing I have to mention is that, in the code, I use STL map template, I
guess STL could be a suspect. I am thinking maybe my compiling line should
do something for STL?

Yes, probably: rwstd points to Rogue Wave and rb_tree to a red-black
tree inside map.

 

[modemer]

Maybe generated template code is stored in separate files, which are
re-generated when the compiler deems so?? Look around, do you see
suspicious files?

Normally Sun Workshop creates a subdirectory "SunWS_cache" in the current
directory. After I removed this subdirectory, the problem is gone, but the
problem did come back when I touched one of the source file and rebuilt. The
only way I can image is that this subdirectory is refreshed correctly.

Yes, probably: rwstd points to Rogue Wave and rb_tree to a red-black
tree inside map.

Panjandrum, thank you very much for your help!

M.