GNU Make Skipping Straight to Linking - makefile

I have a makefile that for various reasons relies on a supporting python script to run every time and grab files from several external locations, copy into working directory, and run through a separate preprocessor before compiling.
This makefile must be able to be run in parallel (-j8) so the order of processing cannot be guaranteed.
In trying to explicitly specify prerequisites, I have created a situation where make skips all object files, goes straight to linking, and fails because the necessary objects do not exist. On a second run, all the objects already exist (the preprocess step skips the files that already exist) and all the files are compiled and linked properly.
When run without -j# everything works fine, but the moment I add -j2, the skipping begins.
Following is an example make file:
GEN_FILES := file1.cpp file2.cpp file3.cpp
CXX_FILES := bin_main.cpp $(GEN_FILES)
OBJ_FILES := $(patsubst %.cpp,%.o,$(CXX_FILES))
.PHONY : all clean prepare
all : bin_file
prepare :
# Copy and preprocess all source files
[ -f file1.cpp ] || cp d1/file1.cpp .
[ -f file2.cpp ] || cp d2/file2.cpp .
[ -f file3.cpp ] || cp d3/file3.cpp .
$(OBJ_FILES) : prepare
bin_file : $(OBJ_FILES)
[ -f file1.o ] && [ -f file2.o ] && [ -f file3.o ] && touch bin_file
%.o : %.cpp
#echo "Compiling $<..."
[ -f $< ] && touch $#
clean :
$(RM) *.o
$(RM) file*
$(RM) bin_file
How can I get this to build in one go, first running prepare to collect all files and then compiling and linking as necessary?

As code_fodder mentions the issue is the creation of the source files.
Basically what happens is, you have not told make how to create these source files, so as far as make knows they don't exist and there's no way to create them. So when make wants to build, for example, file1.o it looks at your pattern rule and finds it could build file1.o from file1.cpp. So then it looks for how to build file1.cpp. No file1.cpp exists, and there is no rule that make knows of that will build it, so make ignores that pattern rule as not matching.
Then make sees the target:
$(OBJ_FILES) : prepare
so it thinks there's no recipe needed to create the object files, and just runs the link line. The next time through, make sees the prepared source files (from the previous build) and then it can use your pattern rule.
If you change your pattern rule to a static pattern rule, where you explicitly tell make exactly what rule to use instead of providing it with a possible rule to use that it can ignore if it doesn't match (which is what a pattern rule is), you'll see the error:
$(OBJ_FILES): %.o : %.cpp
#echo "Compiling $<..."
sleep 1
[ -f $< ] && touch $#
will tell you:
make: *** No rule to make target 'file1.cpp', needed by 'file1.o'. Stop.
Remember, make is looking for a matching pattern rule BEFORE it actually builds anything: it doesn't want to build every possible prerequisite of every possible matching pattern rule, to decide whether or not at the end of it the rule can be used. The rule is matched based on the current state of the filesystem plus rules you have given make about changes it could make. Make has no idea that if it were to invoke the prepare target the source files it was looking for would magically come into existence.
Your basic problem is that this statement is the wrong dependency relationship:
$(OBJ_FILES) : prepare
It's not really true that the object files depend on prepare; what's true is that the PREPARED SOURCE FILES depend on prepare. The object files depend only the "prepared" source files, as your pattern rules shows. This rule should be written, instead:
$(GEN_FILES): prepare
If you do this with -j everything will wait as you want.

Yeah, this gets messy / difficult. The problem you have is that you can specify prerequisite lists - that can work in order, but as soon as you start to use -j then make can start processing prerequisites in any old order. So bin_file requires $(OBJ_FILES) which require prepare. Then %.o requires the same named %.cpp file - which it can do for main.o, but not the filex.o since they don't exist yet - but it tries anyway and fails - in the mean time make (in parallel) is potentially starting to generate the .cpp files, but by this time its too late...etc...
My Prerequisites Build Pattern
I use a very specific prerequisites pattern of my own design - some might frown upon - but I have carefully considered this over the years and found it to be optimal for me.
I create a rule called build or something - which requires build_prerequisites target and then calls make to do the actual build once this is complete:
.PHONY: build
build: build_prerequisites
build:
#echo "start_build"
#$(MAKE) bin_file
This means that build_prerequisites is always run first before the recipe runs. You cant seem to achieve the same forcing of order (at least not easily) using just dependencies. I.e. a list of dependencies can be run in any order with -j, but the rule recipe is always run last.
Now we have this pattern we can fill in the rest. First the build_prerequisites target which does your file generation - I am using echo in my example because I don't have your python script:
.PHONY: build_prerequisites
build_prerequisites:
#echo "build_prerequisites"
echo "create file1" > file1.cpp
echo "create file2" > file2.cpp
echo "create file3" > file3.cpp
Finally add in the c++ compile and link stages - these will be run with the single recursive make call from build - i.e. $(MAKE) bin_file (again I am using echo to create the files in my example):
%.o : %.cpp
#echo "compiling: $<"
##echo "$(CXX) $(SRC_INCLUDES) $(LIB_INCLUDES) $(CXXFLAGS) -c $< -o $#"
#echo "touch" > $#
bin_file : $(OBJ_FILES)
#echo "linking: $<"
#echo $(CXX) $(SRC_INCLUDES) $^ $(LIB_INCLUDES) $(LDFLAGS) -o $#
#echo "touch" > $#
Output
Here is the output from my test program (using echo) and main.cpp already exists usingn -j10:
make -j10
build_prerequisites
echo "create file1" > file1.cpp
echo "create file2" > file2.cpp
echo "create file3" > file3.cpp
start_build
make[1]: Entering directory '/mnt/d/software/ubuntu/make'
compile: bin_main.cpp
compile: file1.cpp
compile: file2.cpp
compile: file3.cpp
link: bin_main.o
g++ bin_main.o file1.o file2.o file3.o -o bin_file
make[1]: Leaving directory '/mnt/d/software/ubuntu/make'
Note: if I put a sleep 1 in the "compile" rule - this still takes only 1 second for all 4 files to compile.
Put it all together
GEN_FILES := file1.cpp file2.cpp file3.cpp
CXX_FILES := bin_main.cpp $(GEN_FILES)
OBJ_FILES := $(patsubst %.cpp,%.o,$(CXX_FILES))
###### STAGE 1
.PHONY: build
build: build_prerequisites
build:
#echo "start_build"
#$(MAKE) bin_file
.PHONY: build_prerequisites
build_prerequisites:
#echo "build_prerequisites"
copy_and_pp_files.py $(CXX_FILES) $(SEARCH_DIRS) .
copy_and_pp_files.py $(CFG_FILES) $(SEARCH_DIRS) .
###### STAGE 2
%.o : %.cpp
#echo "compiling: $<"
#$(CXX) $(SRC_INCLUDES) $(LIB_INCLUDES) $(CXXFLAGS) -c $< -o $#
bin_file : $(OBJ_FILES)
#echo "linking: $<"
#$(CXX) $(SRC_INCLUDES) $^ $(LIB_INCLUDES) $(LDFLAGS) -o $#
###### OTHER RULES
.PHONY: clean
clean :
#$(RM) *.o
#$(RM) file*
I have attempted to use your actual code, but I have no way to test this so there may be a bug in there. I split it up into 2 "stages" for clarity. Stage 1 is done in your makeor make build call, then state 2 is done in the recursive make call in the build recipe.

Related

Make rebuilds everytime

I have a Makefile as :
BUILD_DIR= $(BASE_DIR)/build
_OBJ := a.o b.o
CLEAN_OBJECTS := $(_OBJ)
.PHONY: clean
create_code:
python ../script/my_script.py
all: create_code $(_OBJ)
$(_OBJ): %.o: %.c
mkdir -p $(BUILD_DIR)
$(CC) $(CFLAGS) $(INCLUDE_PATH) -c $< -o $#
cp *.o $(BUILD_DIR)
clean:
rm -f $(CLEAN_OBJECTS)
The target create_code executes the python script and generates a set of .c/.h files.
The target _obj compiles them.
Every time I run make all , create_code target is run even though there is no change to .c/.h generated earlier .
Any suggestions on why this is happening and how to make this create_code target run only if make clean was done earlier .
The underlying problem is that you have one or more generated files that depend on something other than the underlying file system -- namely the contents of your database.
One possibility would be to take advantage of the fact that make, having invoked a rule to rebuild a target will, nonetheless, always check the update time of that target when it is specified as a prerequisite in any other rule.
So, given the rule (untested)...
.PHONY: FORCE
%.c: FORCE
command-to-generate-source $#.tmp
diff -q $#.tmp $# || cp $#.tmp $#
Invoking make foo.c from the command line. make will run the commands...
command-to-generate-source foo.c.tmp
diff -q foo.c.tmp foo.c || cp foo.c.tmp foo.c
where command-to-generate-source foo.c.tmp is expected to leave its output in foo.c.tmp. If the newly generated output file is different than the existing file the cp operation will be run and, hence, the target timestamp will be updated and anything dependent on the target will be updated accordingly.
If, however, the newly generated output file is the same as the existing one then no cp will be run, the target file will be left untouched and make will not consider it to be changed when it appears as a prerequisite in other rules.
This is just one possibility but it's the obvious one given that you already have most (if not all) of the required logic in the command python ../script/my_script.py.

Makefile stops running at the middle [duplicate]

Hopefully this is a very simple question. I have a makefile pattern rule that looks like this:
%.so : %.f %.pyf
f2py -c -L${LAPACK_DIR} ${GRASPLIBS} -m $* $^ ${SOURCES} --opt='-02' --f77flags='-fcray-pointer' >> silent.txt
I want the makefile to build a number of .so files, so I tried to get it to build two files (radgrd_py.so and lodiso_py.so) by doing this:
radgrd_py.so lodiso_py.so:
%.so : %.f %.pyf
f2py -c -L${LAPACK_DIR} ${GRASPLIBS} -m $* $^ ${SOURCES} --opt='-02' --f77flags='-fcray-pointer' >> silent.txt
and then tried this:
radgrd_py.so:
lodiso_py.so:
%.so : %.f %.pyf
f2py -c -L${LAPACK_DIR} ${GRASPLIBS} -m $* $^ ${SOURCES} --opt='-02' --f77flags='-fcray-pointer' >> silent.txt
But in each case, it only builds the first target that I specify. If I run 'make radgrd_py.so' it works fine, I'm just not sure how to specify a list of files that need to be built so I can just run 'make'.
The usual trick is to add a 'dummy' target as the first that depends on all targets you want to build when running a plain make:
all: radgrd_py.so lodiso_py.so
It is a convention to call this target 'all' or 'default'. For extra correctness, let make know that this is not a real file by adding this line to your Makefile:
.PHONY: all
Best way is to add:
.PHONY: all
.DEFAULT: all
all: radgrd_py.so lodiso_py.so
Explanations:
make uses the first target appearing when no .DEFAULT is specified.
.PHONY informs make that the targets (a coma-separated list, in fact) don't create any file or folder.
all: as proposed by schot

Prerequisite `myphony` of target `mytarget` does not exist when rebuilding - as if myphony was a file

My Makefile has been recompiling all my source files even if they haven't changed. I gave it a look with make -d and found stuff like this:
...
Prerequisite `setup' of target `bin/exception/Exceptions.o' does not exist.
Must remake target `bin/exception/Exceptions.o'.
...
Please note that this is what I get after bin/exception/Exceptions.o has already been compiled.
Now what on Earth does it mean by the prerequisite not existing? I've quite clearly declared it, and marked it as PHONY. This example should produce the same results for you (You should just be able to run this - I've made it produce all the necessary files for you)
OBJ_DIR=bin
SRC_DIR=src
OBJS=$(OBJ_DIR)/exception/Exception.o
.PHONY: all
all: $(OBJ_DIR)/app
.PHONY: setup
setup:
mkdir -p $(sort $(dir $(OBJS)))
# To make the source file for you
$(SRC_DIR)/%.cpp:
#mkdir -p $(dir $#)
touch $#
$(OBJ_DIR)/app: setup $(OBJS)
#echo CXX -o $#
#touch $#
$(OBJ_DIR)/%.o: $(SRC_DIR)/%.cpp setup
#echo CXX -c -o $#
#cp $< $#
If I take out the requirement for setup, then my program will not recompile unchanged sources as expected.
Why does make think it needs to rebuild?
[Phony targets] should not be [...] prerequisite[s] of a real target file.
That's from the manual section on Phony targets.
What you are seeing is part of why. make always considers a phony target as out-of-date and needing to be rebuilt. As a result (and since the file doesn't exist) that also applies to anything that depends on it.
There are two simple solutions to this problem.
The first is to make setup a Force Target instead of a .PHONY target which will allow it to operate normally (it might want to also be an Empty Target too but as long as it is touched at least once it should work).
The second is to make setup an order-only prerequisite:
$(OBJ_DIR)/app: $(OBJS) | setup
and
$(OBJ_DIR)/%.o: $(SRC_DIR)/%.cpp | setup
You could also just make sure that the normal output target recipes create the output directory before trying to use it.
$(OBJ_DIR)/app: $(OBJS)
#mkdir -p $(#D)
#echo CXX -o $#
#touch $#
$(OBJ_DIR)/%.o: $(SRC_DIR)/%.cpp
#mkdir -p $(#D)
#echo CXX -c -o $#
#cp $< $#

Calling other makefile from make

I have seen this question and am somehow still unable to get make to properly call another Makefile. Here is my Makefile:
base:
cd Base && $(MAKE)
Base is a sub-directory of my current directory where there is a different Makefile used to compile the files in said folder. However, when I call make, I simply get nothing to be done for base, whereas when I do cd Base && make from the command-line, I get expected behavior (i.e., compilation occurs). In case it is something wrong with the Makefile in Base, here is a copy of that as well:
CC=g++
CFLAGS=-O3 -Wall -pedantic -Werror -c
LINK=g++
LFLAGS=-O3
SRC=main.cpp
OBJ=..\lib\main.o
EXE=..\bin\test.exe
all: $(EXE)
$(EXE): $(OBJ) $(SRC)
$(LINK) $(LFLAGS) $(OBJ) -o $(EXE)
..\lib\main.o: main.cpp
$(CC) $(CFLAGS) $< -o $#
I'm not sure what $(MAKE) amounts to (I'm assuming just 'make'), but I have my makefiles setup in a cascading manner where one calls another in order to cover all subdirectories.
Each makefile has some common targets like 'all' and 'clean'.
Then my makefiles which are just delegating to subdirs look something like this:
.PHONY : all
all:
cd SubDir1 && make $#
cd SubDir2 && make $#
...
.PHONY : clean
clean:
cd SubDir1 && make $#
cd SubDir2 && make $#
...
$# is an automatic variable which will be substituted for the target ('all' or 'clean' in the example above).
This allows you to run 'make', 'make clean', 'make all', or even 'make clean all' from the top-level dir and it will recurse through the subdirs specified in the makefiles for the appropriate target. So if you run 'make clean' at the top, all directories in the makefile's list will be called with 'make clean'.
You can still use this technique even if you don't have common targets. Simply change the automatic variable $# with the appropriate target for each directory's makefile.

Adding dependencies changes the target to the first C file

Consider the following makefile:
.SUFFIXES:
SRC:=../Src
OBJ:=../Obj
# Sources
SOURCES := $(SRC)/App/a.c $(SRC)/App/b.c $(SRC)/App/c.c
HEADERS := $(wildcard $(SRC)/App/*.h)
# Directories
INC_DIRS := $(SRC)/App
OBJ_INC_DIRS := $(INC_DIRS:$(SRC)/%=$(OBJ)/%)
# Objects
OBJECTS := $(SOURCES:$(SRC)%=$(OBJ)%.obj)
# Dependencies
DEPS := $(SOURCES:$(SRC)%.c=$(OBJ)%.d)
-include $(DEPS)
GCC_INCLUDES := $(foreach directory, $(INC_DIRS), -I$(directory))
all: target
target: $(OBJECTS)
touch target
#Objects
$(OBJ)%.c.obj: $(SRC)%.c
#echo Compiling $#
#touch $#
# Dependencies
$(OBJ)%.d: $(SRC)%.c
#echo Checking dependencies for $<
#gcc -MM $< $(GCC_INCLUDES) -MT '$(patsubst %.d,%.c.obj,$#)' -MT '$#' -MF '$#'
#[ ! -s $# ] && rm -f $#
# Creating directory tree before checking dependencies
$(DEPS):|$(OBJ_INC_DIRS)
$(OBJ_INC_DIRS):
#mkdir $#
clean:
echo clean
#rm $(OBJ_INC_DIRS)
When running the first time, I get:
Checking dependencies for ../Src/App/a.c
Checking dependencies for ../Src/App/b.c
Checking dependencies for ../Src/App/c.c
clean
Compiling ../Obj/App/a.c.obj
Compiling ../Obj/App/b.c.obj
Compiling ../Obj/App/c.c.obj
touch target
It's ok, but now, make again (without modifying any file):
make: `../Obj/App/a.c.obj' is up to date.
Now if I modify the file a.c
Checking dependencies for ../Src/App/a.c
Compiling ../Obj/App/a.c.obj
target isn't remade !
It's like my file a.c is the target but it isn't... Can someone explain me what's wrong here?
If I remove the include to the DEPS, I observe the expected behavior...
Thanks
EDIT
By putting the include at the end as mentioned by #Beta works but now I added the target clean and show the result...
I'll have to do some experiments to be sure, but I think the problem is:
-include $(DEPS)
...
all: target
You include $(DEPS) before the first target. So if you modify a.c, Make sees that it must rebuild a.d, then since it includes that file it must start over, and now a.c.obj is an earlier target than all.
Try moving -include $(DEPS) to the end of the makefile.
EDIT:
(Two small points: your clean rule is incorrect, since it tries to rm a directory, and I would do make clean; make all rather than make all, since I am not certain that Make promises to build targets in the given order in all cases.)
Yes, this makefile will rebuild the DEPS even when running clean. The makefile includes those files and has a rule for them, so if they are missing or out of date it must rebuild them and restart, no matter what the target is. The best way to deal with this is by Advanced Auto-Dependency Generation; basically, the commands that build dependency files go in the %.obj rule, so that a.d is a side effect of building a.c.obj. It's a sophisticated technique, not obvious, but it works beautifully. (Let us know if you try this and have trouble setting it up.)

Resources