I pulled a cvs update for the xtrn directory about an hour ago. I noticed that the dpoker code had been updated so I attempted a make in that directory:
dpoker.c:1712:29: note: ...this statement, but the latter is misleadingly indented as if it is guarded by the æifÆ
if (m==j) j++; if (m1==j) j++; if (m2==j) j++;
^~
../../src/build/Common.gmake:457: recipe for target 'gcc.linux.x64.obj.debug-mt/dpoker.o' failed
make: *** [gcc.linux.x64.obj.debug-mt/dpoker.o] Error 1
I think I know what is wrong... /xtrn/dpoker/gcc.linux.x64.obj.debug-mt is an empty directory. the dpoker.o file appears to exist in the obj.release-mt directory. Copying the members from obj.release-mt to obj.debug-mt does not fix the issue. The error remains the same.
../../src/build/Common.gmake:457: recipe for target 'gcc.linux.x64.obj.debug-mt/dpoker.o' failedThe actual error message should be further up. In any case, try a "make clean" >irst.
make: *** [gcc.linux.x64.obj.debug-mt/dpoker.o] Error 1
I think I know what is wrong... /xtrn/dpoker/gcc.linux.x64.obj.debug-mt is an empty directory. the dpoker.o file appears to exist in the obj.release-mt directory. Copying the members from obj.release-mt to obj.debug-mt does not fix the issue. The error remains the same.
../../src/build/Common.gmake:457: recipe for target 'gcc.linux.x64.obj.debug-mt/dpoker.o' failedThe actual error message should be further up. In any case, try a "make clean" >irst.
make: *** [gcc.linux.x64.obj.debug-mt/dpoker.o] Error 1
Would a make clean fix a directory being empty, as per my findings below?
Or is the .o file something the compile would build if it had worked?
Unlikely. A "make clean" will remove files, not add them.
Unlikely. A "make clean" will remove files, not add them.
OK so I ran the ./cleanall.sh RELEASE=1
Then I ran the recompile script that included this:
export CVSROOT=:pserver:anonymous@cvs.synchro.net:/cvsroot/sbbs
cvs update -d src 3rdp
I did not receive any errors recompiling.
I then changed to the xtrn/dpoker directory and ran:
make clean
make
It looks like something is up with conio:
make -C ../../src/conio mtlib
make[1]: Entering directory '/sbbs/src/conio'
Compiling ansi_cio.c
Compiling ciolib.c
make[1]: *** No rule to make target '../smblib/base64.h', needed by 'gcc.linux.x64.obj.debug-mt/cterm.o'. Stop.
make[1]: Leaving directory '/sbbs/src/conio' ../../src/build/Common.gmake:505: recipe for target 'ciolib-mt' failed
make: *** [ciolib-mt] Error 2
OK so I ran the ./cleanall.sh RELEASE=1
Okay, so that would clean all the "release" binaries, but leave any "debug" bin
ries alone.
make clean
make
It appears you just ran "make" here, which is "debug" build by default. And you
have an old dependency stored for the old locatino of base64.h. Running "cleana
l.sh" without the "RELEASE=1" argument, first, should take care of that.
OK so I ran the ./cleanall.sh RELEASE=1
Okay, so that would clean all the "release" binaries, but leave any "debug" bin
ries alone.
make clean
make
It appears you just ran "make" here, which is "debug" build by default. And you
have an old dependency stored for the old locatino of base64.h. Running "cleana
l.sh" without the "RELEASE=1" argument, first, should take care of that.
If I run cleanall.sh again, do I also need to recompile synchronet after doing so?
When running a make to compile one of the doors, is there
something I should put on the command line that will make it compile the non-debug version (like RELEASE=1)?
When running a make to compile one of the doors, is there
something I should put on the command line that will make it compile the non-debug version (like RELEASE=1)?
Yes, 'make RELEASE=1' to make a release binary, even for dpoker.
Sysop: | Kurt Hamm |
---|---|
Location: | Columbia, SC |
Users: | 8 |
Nodes: | 20 (0 / 20) |
Uptime: | 136:05:38 |
Calls: | 2,804 |
Calls today: | 1 |
Files: | 64 |
Messages: | 854,509 |