VERS := naive DC := dmd DL := dmd GNUPLOT := gnuplot DFLAGS += -defaultlib=tango-base-dmd-$(VERS) -debuglib=tango-base-dmd-$(VERS) DFLAGS += -release -inline -O O := $(VERS) ifndef V P = @ P_DC = @printf ' DC %- 40s <- %s\n' '$@' '$(if \ $(subst 1,,$(words $^)),$< ... $(lastword $^),$<)'; P_PLOT = @printf ' PLOT %- 40s <- %s\n' '$@' '$(filter %.csv,$^)'; P_AWK = @printf ' AWK %- 40s <- %s\n' '$@' '$<'; P_RUN = @printf ' RUN $< $(args)\n'; P_MAKE = @printf ' MAKE $@\n'; P_RM = @printf ' RM $^\n'; endif # create build directories if they don't already exist ifneq ($(wildcard $O),$O) dummy_mkdir := $(shell mkdir -p $O) endif # don't use Gold with old DMDs ifneq ($(shell ld --version | grep gold),) export LD_ := /usr/bin/ld.single endif .PHONY: all all: naive .PHONY: naive basic naive basic: $(P_MAKE) $(MAKE) --no-print-directory micro VERS=$@ # micro ######## .PHONY: build-micro build-micro: $(patsubst %.d,$O/%,$(wildcard micro/*.d)) .PHONY: micro micro: $(patsubst %.d,$O/%.eps,$(wildcard micro/*.d)) # special command line arguments 'split' micro benchmark $O/micro/split.c.csv $O/micro/split.a.csv: override args := micro/bible.txt # common rules ############### .PRECIOUS: $O/% $O/%: %.d $(P_DC) $(DC) $(DFLAGS) -of$@ $^ .PRECIOUS: $O/%.c.csv $O/%.a.csv $O/%.c.csv $O/%.a.csv: $O/% $(P_RUN) D_GC_STATS=1 ./$< $(args) $P mv gc-collections.csv $O/$*.c.csv $P mv gc-mallocs.csv $O/$*.a.csv .PRECIOUS: $O/%.h.csv $O/%.h.csv: $O/%.a.csv hist.awk $(P_AWK) awk -F, -f $(lastword $^) $< > $@ .PRECIOUS: $O/%.tics $O/%.tics: $O/%.h.csv tics.awk $(P_AWK) awk -F, -f $(lastword $^) $< > $@ $O/%.eps: $O/%.c.csv $O/%.a.csv $O/%.h.csv $O/%.tics plot.gpi $(P_PLOT) sed "s|@@PRG@@|$(*F)|g; s|@@COL@@|$(VERS)|g; \ s|@@INC@@|$(word 1,$^)|g; s|@@INA@@|$(word 2,$^)|g; \ s|@@INH@@|$(word 3,$^)|g; s|@@OUT@@|$@|g; \ s|@@TICS@@|$(shell cat $(word 4,$^))|g" $(word 5,$^) \ | $(GNUPLOT) .PHONY: clean clean: $O/ $(P_RM) $(RM) -r $^