This Time Self-Hosted
dark mode light mode Search

Tell-tale signs that your Makefile is broken

Last week I sent out a broad last-rite email for a number of gkrellm plugins that my tinderbox reported warnings about that shows that they have been broken for a long time. This has been particularly critical because the current maintainer of all the gkrellm packages, Jim (lack), seems not to be very active on them.

The plugins I scheduled for removal are mostly showing warnings related to the gdk_string_width() function called with a completely different object than it should have been called with, which will result in unpredictable behaviour at runtime (most likely, it’ll crash). A few more were actually buffer overflows, or packages failing because their dependencies changed. If you care about a plugin that is scheduled for removal, you’re suggested to look into it yourself and start proxy-maintain it.

I originally though I was able to catch all of the broken packages; but since then, another one appeared with the same gdk_string_width() error, so I decided running the tinderbox specifically against the gkrellm plugins; there was another one missing and then I actually found all of them. A few more were reported ignoring LDFLAGS, but nothing especially bad turned up on my tinderbox.

What it did show though, is that the ignored LDFLAGS are just a symptom of a deeper problem: most of the plugins have broken Makefile that are very poorly written. This could be seen in a number of small things, but the obvious one is the usual ”job server unavailable” message that I have written about last year.

So here’s a good checklist of things that shows that your Makefile is broken:

  • you call directly the make command — while this works perfectly fine on GNU systems, where you almost always use the GNU make implementation, this is not the case in most BSD systems, and almost always the Makefile is good enough only to work with the GNU implemenation; the solution is to call $(MAKE) which is replaced with the name of the make implementation you’re actually using;
  • it takes you more than one command to run make in a subdirectory (this can also be true for ebuilds, mind you) — things like cd foo && make or even worse (cd foo; make; cd ..; ) are mostly silly to look at and, besides, will cause the usual jobserver unavailable warning; what you might not know here is that make is (unfortunately) designed to allow for recursive build, and provides an option to do so without requiring changing the working directory beforehand: make -C foo (which actually should be, taking the previous point into consideration, $(MAKE) -C foo) does just that, and only changes the working directory for the make process and its children rather than for the current process as well;
  • it doesn’t use the builtin rules — why keep writing the same rules to build object files? make already knows how to compile .c files into relocatable objects; instead of writing your rules to inject parameters, just use the CFLAGS variable like make is designed to do! Bonus points if, for final executables, you also use the built-in linking rule (for shared objects I don’t think there is one);
  • it doesn’t use the “standard” variable names — for years I have seen projects written in C++ insisting on using CPP and CPPFLAGS variables, well that’s wrong, as here “cpp” refers to the C Pre-Processor; the correct variables are CXX and CXXFLAGS; inventing your own variable names to express parameters that can be passed by the user tends to be a vary bad choice, as you break the expectations of the developers and packagers using your software.

Now, taking this into consideration, can you please clean up your packages? Pretty please with sugar on top?

Comments 13
  1. I am certainly not an expert, but I think the “using builtin rules” isn’t always such a great idea, e.g. you might want to customize the output (e.g. kernel style), you need some parts to be compiled with a cross-compiler and others with the host compiler which might make the builtin rules confusing.But if you don’t use (some of) the builtin (suffix-)rules I suggest you disable them, that speeds up processing quite a bit, particularly on Windows and seems to avoid some cases where make confuses itself and fails to compile (compiling .rc files with some make versions somehow manage to convince make that everything is up-to-date despite the destination file not even existing).

  2. Obviously, the usual rule of “if you do know what you’re doing, keep doing it” applies.The problem is that _most_ people who don’t use the builtin rules will get them wrong in one way or another.

  3. Saw a great example of that final point the other day. D2X-XL uses the ARCH variable for arch-specific CFLAGS. Obviously it didn’t agree with Portage!

  4. Right, without subshell it works; on the other hand, why using two commands when one suffices? 🙂

  5. Another point about recursion. As far as I understand “$(MAKE) -C” is a gnu-ism.With that in mind I was thinking it was better to go with “cd subdir && $(MAKE)” if you intended to support a wider range of make utilities.

  6. Might have to look deeper into that later on today, but I’m pretty sure that @$(MAKE) -C@ is quite standard, as at least BSD supports it.

  7. Actually, it can by now be a good option to simply rely on GNU make instead of a generic make, because it can save you so many other dependencies that would be needed to work around broken or incomplete implementations (Perl etc.). Also in a modern build one should never need to recurse, but have one build process handling all the dependencies.(1)<shameless plug>see the new OpenOffice.org build system for example:http://www.ooocon.org/index…</shameless>(1) see also: http://miller.emu.id.au/pmi

  8. I don’t know if FreeBSD supports $(MAKE) -C, but I do know that NetBSD does not. From what I understand, the officially correct and portable way to recurse actually is “cd dir && $(MAKE)”. But I agree with Peter Miller who was just referenced by Bjoern recursive make should definitely be avoided for reasons that Diego has gone over many times.

  9. I cannot watch it right now but I’ll make sure to watch it tomorrow, since it really is in my area of interest, thanks!

  10. What we need here is more discussion on make -C! According to the online man pages, (Free|Dragonfly|Net)BSD all support it, while OpenBSD does not. Though I never find the need to make recursive builds, I will do it the non-GNU way. GNU has too much glut. Hmmm, maybe ‘Glut, Not Usability’.I was looking at some broken make files today, and noticed some CPPFLAGS garbage, CC=g++, a static archive depending directly on headers…After an overhaul, writing ‘make’ actually uses dependency information correctly.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.