On 10.12.2025 18:08, Jason Andryuk wrote: > On 2025-12-10 09:40, Anthony PERARD wrote: >> On Tue, Dec 09, 2025 at 04:47:28PM -0500, Jason Andryuk wrote: >>> diff --git a/xen/Makefile b/xen/Makefile >>> index e6cf287425..aeb5dcf2ee 100644 >>> --- a/xen/Makefile >>> +++ b/xen/Makefile >>> @@ -469,10 +469,13 @@ all-symbols-$(CONFIG_FAST_SYMBOL_LOOKUP) += >>> --sort-by-name >>> >>> include $(srctree)/arch/$(SRCARCH)/arch.mk >>> >>> +XEN_FINAL_LDFLAGS-$(CONFIG_GC_SECTIONS) := --gc-sections >> >> Is there a good reason to add this flags after the arch-specific >> makefiles? If not, could you move that just before, and right after the >> definition of "$(all-symbols)" as it's a variable that is used in the >> same phase of the build. (With Jan's other feedback) > > No, there is no reason for its location. I can move it. > >>> # define new variables to avoid the ones defined in Config.mk >>> export XEN_CFLAGS := $(CFLAGS) >>> export XEN_AFLAGS := $(AFLAGS) >>> export XEN_LDFLAGS := $(LDFLAGS) >>> +export XEN_FINAL_LDFLAGS := $(LDFLAGS) $(XEN_FINAL_LDFLAGS-y) >> >> "FINAL" isn't very descriptive. A completely wrong interpretation might >> be that we should use the "final" variable instead of "XEN_LDFLAGS". How >> about a name that describe where this set of flags is going to be used, >> like "XEN_LDFLAGS_xen_syms" (which unfortunately doesn't exactly fit >> with x86 xen.efi target), or maybe suffix it with "_target" or just >> "_xen"? (In Linux build system, they use "LDFLAGS_vmlinux", but I don't >> know what would be the equivalent of "vmlinux" in our build system.) > > I plan to use "_xen" unless anyone objects. "_xen_lds" could be another > option, but again that doesn't match efi.lds.
_lds would also be wrong - that rather refers to the linker script than the final binary linking of which these flags influence. Jan
