Re: [PATCH] vfio: selftests: Fix out-of-tree build with make O=

[David Matlack <dmatlack@google.com>]

On 2026-04-28 04:04 PM, Jason Gunthorpe wrote:
> On Tue, Apr 28, 2026 at 06:27:06PM +0000, David Matlack wrote:
> > On 2026-04-27 08:07 PM, Jason Gunthorpe wrote:
> > > The test programs are compiled via a static pattern rule that requires
> > > intermediate .o files:
> > > 
> > >   $(TEST_GEN_PROGS): %: %.o $(LIBVFIO_O)
> > > 
> > > After lib.mk prefixes TEST_GEN_PROGS with $(OUTPUT), this creates
> > > dependencies on .o files in the output directory (e.g.
> > > $(OUTPUT)/vfio_dma_mapping_test.o). However, there is no rule to compile
> > > these .o files from the source directory .c files when OUTPUT differs
> > > from the source directory.
> > > 
> > > Add an explicit chain of pattern rules:
> > >   $(OUTPUT)/% -> $(OUTPUT)/%.o -> %.c
> > > 
> > > Following the same pattern already used in libvfio.mk for the library
> > > objects.
> > > 
> > > Fixes: 19faf6fd969c ("vfio: selftests: Add a helper library for VFIO selftests")
> > > Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

With the fix applied from your reply,

Reviewed-by: David Matlack <dmatlack@google.com>

> > > ---
> > >  tools/testing/selftests/vfio/Makefile | 5 ++++-
> > >  1 file changed, 4 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/tools/testing/selftests/vfio/Makefile b/tools/testing/selftests/vfio/Makefile
> > > index 40165d087a0bc4..81c3be8298fdc2 100644
> > > --- a/tools/testing/selftests/vfio/Makefile
> > > +++ b/tools/testing/selftests/vfio/Makefile
> > > @@ -27,10 +27,13 @@ CFLAGS += $(EXTRA_CFLAGS)
> > >  
> > >  LDFLAGS += -pthread
> > >  
> > > -$(TEST_GEN_PROGS): %: %.o $(LIBVFIO_O)
> > > +$(TEST_GEN_PROGS): $(OUTPUT)/%: %(OUTPUT)/%.o $(LIBVFIO_O)
> > 
> > Changing this line (with the fix you replied with) is not actually
> > needed right? I tried the following without it and was able to build:
> > 
> >   $ make OUTPUT=/tmp -C tools/testing/selftests/vfio
> > 
> > $(TEST_GEN_PROGS) has already been prefixed with $(OUTPUT), so %
> > in the original pattern already included $(OUTPUT).
> > 
> > Is this for readability?
> 
> Yes, it makes more sense to explicitly match the stems through the
> flow than to have the implicit $(TEST_GEN_PROGS) == XXX/$(OUTPUT)

Sounds good.

> 
> > >  	$(CC) $(CFLAGS) $(CPPFLAGS) $(LDFLAGS) $< $(LIBVFIO_O) $(LDLIBS) -o $@
> > >  
> > >  TEST_GEN_PROGS_O = $(patsubst %, %.o, $(TEST_GEN_PROGS))
> > > +$(TEST_GEN_PROGS_O): $(OUTPUT)/%.o: %.c
> > > +	$(CC) $(CFLAGS) $(CPPFLAGS) $(TARGET_ARCH) -c $< -o $@
> > 
> > Looking at this commit made me confused how this was even working before
> > and I learned that make has built-in rules to build .o files that
> > assumes the .c is in the same directory. 
> 
> Yes
> 
> > I see now that we were reying
> > on those built-in rules to build $(TEST_GEN_PROGS_O). It makes sense to
> > have an explicit rule to handle the directory mismatch between .c and
> > .o.
> 
> Hmm now that I think about it, I wonder if it used the right flags..

It looks like it used the same command that we will be using after this
patch, just with the wrong path to the .c file.

 $ make -p -f /dev/null
 ...
 COMPILE.c = $(CC) $(CFLAGS) $(CPPFLAGS) $(TARGET_ARCH) -c
 ...
 OUTPUT_OPTION = -o $@
 ...
  %.o: %.c
  #  recipe to execute (built-in):
          $(COMPILE.c) $(OUTPUT_OPTION) $<