[PATCH] make: don't spray static check failures all over the subdir build

[PATCH] make: don't spray static check failures all over the subdir build

[Darrick J. Wong]

From: Darrick J. Wong <darrick.wong@oracle.com>

Debian package building is special -- it directly calls make -C libxfs
when building the debian-installer packages.  This means that any
variables we define in the top level Makefile don't get passed down to
subdir make processes.

This means that the new static checker support effectively runs the
first argument in $(CFLAGS) as a command, which is surprising.  Fix up
buildrules to patch out CHECK_CMD if nobody's defined it, so that direct
subdir make works again.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 include/buildrules |    4 ++++
 1 file changed, 4 insertions(+)

diff --git a/include/buildrules b/include/buildrules
index 83dfe05b..5c3e6eb9 100644
--- a/include/buildrules
+++ b/include/buildrules
@@ -36,6 +36,10 @@ $(SUBDIRS):
 	$(Q)$(MAKE) $(MAKEOPTS) -q -C $@ || $(MAKE) $(MAKEOPTS) -C $@
 endif
 
+ifndef CHECK_CMD
+CHECK_CMD = @true
+endif
+
 #
 # Standard targets
 #

Re: [PATCH] make: don't spray static check failures all over the subdir build

[Eric Sandeen]

On 11/9/18 1:28 PM, Darrick J. Wong wrote:
> From: Darrick J. Wong <darrick.wong@oracle.com>
> 
> Debian package building is special -- it directly calls make -C libxfs
> when building the debian-installer packages.  This means that any
> variables we define in the top level Makefile don't get passed down to
> subdir make processes.
> 
> This means that the new static checker support effectively runs the
> first argument in $(CFLAGS) as a command, which is surprising.  Fix up
> buildrules to patch out CHECK_CMD if nobody's defined it, so that direct
> subdir make works again.
> 
> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> ---
>  include/buildrules |    4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/include/buildrules b/include/buildrules
> index 83dfe05b..5c3e6eb9 100644
> --- a/include/buildrules
> +++ b/include/buildrules
> @@ -36,6 +36,10 @@ $(SUBDIRS):
>  	$(Q)$(MAKE) $(MAKEOPTS) -q -C $@ || $(MAKE) $(MAKEOPTS) -C $@
>  endif
>  
> +ifndef CHECK_CMD
> +CHECK_CMD = @true
> +endif
> +

Hm, ok, not quite clear if this is a hack or a proper fix - maybe
all the check stuff should have gone into buildrules in the first
place?  But at this late stage in the game, it's probably good enough.

Reviewed-by: Eric Sandeen <sandeen@redhat.com>

Thanks,
-eric

Re: [PATCH] make: don't spray static check failures all over the subdir build

[Darrick J. Wong]

On Fri, Nov 09, 2018 at 01:34:44PM -0600, Eric Sandeen wrote:
> On 11/9/18 1:28 PM, Darrick J. Wong wrote:
> > From: Darrick J. Wong <darrick.wong@oracle.com>
> > 
> > Debian package building is special -- it directly calls make -C libxfs
> > when building the debian-installer packages.  This means that any
> > variables we define in the top level Makefile don't get passed down to
> > subdir make processes.
> > 
> > This means that the new static checker support effectively runs the
> > first argument in $(CFLAGS) as a command, which is surprising.  Fix up
> > buildrules to patch out CHECK_CMD if nobody's defined it, so that direct
> > subdir make works again.
> > 
> > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > ---
> >  include/buildrules |    4 ++++
> >  1 file changed, 4 insertions(+)
> > 
> > diff --git a/include/buildrules b/include/buildrules
> > index 83dfe05b..5c3e6eb9 100644
> > --- a/include/buildrules
> > +++ b/include/buildrules
> > @@ -36,6 +36,10 @@ $(SUBDIRS):
> >  	$(Q)$(MAKE) $(MAKEOPTS) -q -C $@ || $(MAKE) $(MAKEOPTS) -C $@
> >  endif
> >  
> > +ifndef CHECK_CMD
> > +CHECK_CMD = @true
> > +endif
> > +
> 
> Hm, ok, not quite clear if this is a hack or a proper fix - maybe
> all the check stuff should have gone into buildrules in the first
> place?  But at this late stage in the game, it's probably good enough.

It's a stupid hack, but if CFLAGS=-g -O2 -Wall and someone runs "make -C
repair" instead of "make repair" then make will foolishly run "g -O2
-Wall", whatever that does.  command-not-found tells me there's no known
"g" command.

As for debian builds we probably should just update debian/rules to
something more modern.

--D

> Reviewed-by: Eric Sandeen <sandeen@redhat.com>
> 
> Thanks,
> -eric