On 24.08.2021 12:50, Anthony PERARD wrote: > Rather than preparing the efi source file, we will copy them as needed > from the build location. > > Avoid the links as they seems fragile in out-of-tree builds. Also by > making a copy, we don't need to figure out the relative path or we > don't need to use absolute path.
I agree that symlinks wouldn't be nice for the out-of-tree build case. Otoh please see https://sourceware.org/git/?p=binutils-gdb.git;a=commitdiff;h=8c6740616cd244e5763e974cb737affbe71db385 albeit I'll admit the situation was a little different there because it's pre-built files which get populated into the build tree. > --- a/xen/arch/arm/efi/Makefile > +++ b/xen/arch/arm/efi/Makefile > @@ -1,4 +1,10 @@ > CFLAGS-y += -fshort-wchar > +CFLAGS-y += -I$(srctree)/common/efi Perhaps another opportunity for -iquote? > obj-y += boot.init.o pe.init.o ebmalloc.o runtime.o > obj-$(CONFIG_ACPI) += efi-dom0.init.o > + > +$(obj)/%.c: common/efi/%.c > + $(Q)cp -f $< $@ In case both trees are on the same file system, trying to hardlink first would seem desirable. When copying, I think you should also pass -p. > --- a/xen/arch/x86/efi/Makefile > +++ b/xen/arch/x86/efi/Makefile > @@ -1,4 +1,5 @@ > CFLAGS-y += -fshort-wchar > +CFLAGS-y += -I$(srctree)/common/efi > > quiet_cmd_objcopy_o_ihex = OBJCOPY $@ > cmd_objcopy_o_ihex = $(OBJCOPY) -I ihex -O binary $< $@ > @@ -19,3 +20,8 @@ obj-$(XEN_BUILD_EFI) := $(filter-out %.init.o,$(EFIOBJ-y)) > obj-bin-$(XEN_BUILD_EFI) := $(filter %.init.o,$(EFIOBJ-y)) > extra-$(XEN_BUILD_EFI) += buildid.o relocs-dummy.o > nocov-$(XEN_BUILD_EFI) += stub.o > + > +$(obj)/%.c: common/efi/%.c > + $(Q)cp -f $< $@ > + > +.PRECIOUS: $(obj)/%.c Seeing you repeat everything here, despite it not being all this much I wonder if there wouldn't better be a makefile fragment in common/efi/ which all interested architectures' arch/<arch>/efi/Makefile would then include. This could then also subsume -fshort-wchar. Jan
