Re: [PATCH-next v3] arm32: enable HAVE_LD_DEAD_CODE_DATA_ELIMINATION

[Ard Biesheuvel <ardb@kernel.org>]

On Fri, 15 Mar 2024 at 13:10, liuyuntao (F) <liuyuntao12@huawei.com> wrote:
>
>
>
> On 2024/3/15 19:51, Arnd Bergmann wrote:
> > On Fri, Mar 15, 2024, at 12:48, liuyuntao (F) wrote:
> >> On 2024/3/15 18:46, Ard Biesheuvel wrote:
> >>> @@ -28,6 +28,7 @@
> >>>                                      -DEFI_HAVE_MEMCHR -DEFI_HAVE_STRRCHR \
> >>>                                      -DEFI_HAVE_STRCMP -fno-builtin -fpic \
> >>>                                      $(call cc-option,-mno-single-pic-base)
> >>> +cflags-$(CONFIG_ARM)           := $(filter-out -fdata-sections, $(CFLAGS-y))
> >>>    cflags-$(CONFIG_RISCV)         += -fpic -DNO_ALTERNATIVE -mno-relax
> >>>    cflags-$(CONFIG_LOONGARCH)     += -fpie
> >>
> >> Another error:
> >> cannot initialize array of ‘short unsigned int’ from a string literal
> >> with type array of ‘unsigned int’
> >>    17 | static const efi_char16_t shim_MokSBState_name[] = L"MokSBStateRT";
> >
> > The line sets cflags-y to an empty string.
> >
> > It probably should have been
> >
> > cflags-$(CONFIG_ARM)           := $(filter-out -fdata-sections, $(cflags-y))
> >
> > (small cflags instead of CFLAGS).
> >

Oops indeed.

> Oh, my mistake, after fixed this bug, still this error:
> > arm-linux-gnueabi-ld: warning: orphan section `.data.efi_loglevel' from `drivers/firmware/efi/libstub/printk.stub.o' being placed in section `.data.efi_loglevel'
> > arm-linux-gnueabi-ld: warning: orphan section `.data.screen_info_guid' from `drivers/firmware/efi/libstub/screen_info.stub.o' being placed in section `.data.screen_info_guid'
> > arm-linux-gnueabi-ld: warning: orphan section `.data.cpu_state_guid' from `drivers/firmware/efi/libstub/arm32-stub.stub.o' being placed in section `.data.cpu_state_guid'
> > arm-linux-gnueabi-ld: warning: orphan section `.data.efi_nokaslr' from `drivers/firmware/efi/libstub/efi-stub-helper.stub.o' being placed in section `.data.efi_nokaslr'
>
> The -fdata-sections option is present in the KBUILD_CFLAGS_KERNEL
> variable; altering cflags-y only affect the KBUILD_CFLAGS variable.
> It seems that the -fdata-sections option needs to be removed from the
> KBUILD_CFLAGS_KERNEL variable.

Indeed.

So

KBUILD_CFLAGS_KERNEL := $(filter-out -fdata-sections,$(KBUILD_CFLAGS_KERNEL))

(assuming I got it right this time without build testing)