Alex Rousskov wrote:
OK. So there is no mallopt in your preprocessed MemPool.E, as expected.
Let's see whether g++ will suddenly find it if you compile MemPool.E.
Run the same compilation command, but with a -c option at the end and
adjusted output/input file names:
g++ -DHAVE_CONFIG_H -I. -I. -I../include -I../include -I../include -Werror -Wall -Wpointer-arith -Wwrite-strings -Wcomments -D_REENTRANT -g -O2 -MT MemPool.o -MD -MP -MF ".deps/MemPool.Tpo" -c -o MemPool.o MemPool.E
here is the output,i hope i type it correctly
core# g++ -DHAVE_CONFIG_H -I. -I. -I../include -I../include -I../include
-Werror -Wall -Wpointer-arith -Wwrite-strings -Wcomments -D_REENTRANT -g
-O2 -MT MemPool.o -MD -MP -MF ".deps/MemPool.Tpo" -c -o MemPool.o MemPool.E
g++: MemPool.E: linker input file unused because linking not done
core#
If it fails, email me your MemPool.E along with the g++ output.
this is the link to the MemPool.E
http://www.bps.co.id/~zen/MemPool.E
If it succeeds, we would have a case of two pretty much identical
compilations producing different results. If you recall, nearly the same
command failed when you ran "make" (we just ran the preprocessing step
manually). If this happens, try running "
make -k distclean;
./configure ... && make
again.
Thank you,
Alex.
TIA
Zen
