On Thu, Aug 15, 2024 at 2:50 PM Eduard Zingerman <eddyz87@xxxxxxxxx> wrote: > > On Thu, 2024-08-15 at 14:06 -0700, Andrii Nakryiko wrote: > > [...] > > > > @@ -627,6 +669,9 @@ ifneq ($2:$(OUTPUT),:$(shell pwd)) > > > $(Q)rsync -aq $$^ $(TRUNNER_OUTPUT)/ > > > endif > > > > > > +$(OUTPUT)/$(TRUNNER_BINARY): LDLIBS += $$($(TRUNNER_BASE_NAME)-LDLIBS) > > > +$(OUTPUT)/$(TRUNNER_BINARY): LDFLAGS += $$($(TRUNNER_BASE_NAME)-LDFLAGS) > > > > is there any reason why you need to have this blah-LDFLAGS convention > > and then applying that with extra pass, instead of just writing > > > > $(OUTPUT)/$(TRUNNER_BINARY): LDFLAGS += $(LLVM_LDFLAGS) > > > > I'm not sure I understand the need for extra logical hops to do this > > No real reason, that's how it is organized in bpftool makefile, > monkey see, monkey do. Will combine to have single LDFLAGS change. I think such an approach makes sense for Linux kernel where there is a Kbuild system of conventions and you mostly don't write real Makefile statements (just declaratively specifying a few bits here and there). But that's not the case for BPF selftests (and not for bpftool, but that's a separate discussion), so extra hops just make everything harder to follow, IMO. > > [...] >
