| Message ID | 20200626164307.3327380-6-dinechin@redhat.com |
|---|---|
| State | New |
| Headers | show |
| Series | RFC: Move SPICE to a load module | expand |
On Fri, Jun 26, 2020 at 06:43:02PM +0200, Christophe de Dinechin wrote: > Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> > Signed-off-by: Christophe de Dinechin <dinechin@redhat.com> > --- > Makefile.objs | 1 + > Makefile.target | 7 +++++++ > hw/Makefile.objs | 1 + > 3 files changed, 9 insertions(+) > > diff --git a/Makefile.objs b/Makefile.objs > index e38768c8d5..6703353493 100644 > --- a/Makefile.objs > +++ b/Makefile.objs > @@ -86,6 +86,7 @@ endif # CONFIG_SOFTMMU > # Target-independent parts used in system and user emulation > common-obj-y += cpus-common.o > common-obj-y += hw/ > +common-obj-m += hw/ > common-obj-y += qom/ > common-obj-y += disas/ > > diff --git a/Makefile.target b/Makefile.target > index 8ed1eba95b..3f3b5ee058 100644 > --- a/Makefile.target > +++ b/Makefile.target > @@ -179,6 +179,13 @@ endif # CONFIG_SOFTMMU > dummy := $(call unnest-vars,,obj-y) > all-obj-y := $(obj-y) > > +# > +# common-obj-m has some crap here, probably as side effect from > +# filling obj-y. Clear it. Fixes suspicious dependency errors when > +# building devices as modules. > +# > +common-obj-m := This comment doesn't fill me with confidence - makes it feel like there's some more important root cause that needs addressing instead. > include $(SRC_PATH)/Makefile.objs > dummy := $(call unnest-vars,.., \ > authz-obj-y \ > diff --git a/hw/Makefile.objs b/hw/Makefile.objs > index 4cbe5e4e57..d6d387b74b 100644 > --- a/hw/Makefile.objs > +++ b/hw/Makefile.objs > @@ -43,4 +43,5 @@ devices-dirs-y += smbios/ > endif > > common-obj-y += $(devices-dirs-y) > +common-obj-m := display/ > obj-y += $(devices-dirs-y) > -- > 2.26.2 > > Regards, Daniel
Hi, > > +# > > +# common-obj-m has some crap here, probably as side effect from > > +# filling obj-y. Clear it. Fixes suspicious dependency errors when > > +# building devices as modules. > > +# > > +common-obj-m := > > This comment doesn't fill me with confidence - makes it feel like there's > some more important root cause that needs addressing instead. It's my fault, see also https://patchwork.ozlabs.org/project/qemu-devel/patch/20200624131045.14512-5-kraxel@redhat.com/ I think the underlying problem is that the functions building *-m variables do not properly handle per-target objects. Which has no bad side effects as long as we never recurse into hw/, but if we want build devices as modules we have to ... Unless we want build per-target code as module the above should do the trick I think. Better suggestions are welcome of course. take care, Gerd
diff --git a/Makefile.objs b/Makefile.objs index e38768c8d5..6703353493 100644 --- a/Makefile.objs +++ b/Makefile.objs @@ -86,6 +86,7 @@ endif # CONFIG_SOFTMMU # Target-independent parts used in system and user emulation common-obj-y += cpus-common.o common-obj-y += hw/ +common-obj-m += hw/ common-obj-y += qom/ common-obj-y += disas/ diff --git a/Makefile.target b/Makefile.target index 8ed1eba95b..3f3b5ee058 100644 --- a/Makefile.target +++ b/Makefile.target @@ -179,6 +179,13 @@ endif # CONFIG_SOFTMMU dummy := $(call unnest-vars,,obj-y) all-obj-y := $(obj-y) +# +# common-obj-m has some crap here, probably as side effect from +# filling obj-y. Clear it. Fixes suspicious dependency errors when +# building devices as modules. +# +common-obj-m := + include $(SRC_PATH)/Makefile.objs dummy := $(call unnest-vars,.., \ authz-obj-y \ diff --git a/hw/Makefile.objs b/hw/Makefile.objs index 4cbe5e4e57..d6d387b74b 100644 --- a/hw/Makefile.objs +++ b/hw/Makefile.objs @@ -43,4 +43,5 @@ devices-dirs-y += smbios/ endif common-obj-y += $(devices-dirs-y) +common-obj-m := display/ obj-y += $(devices-dirs-y)