On Fri, Mar 12, 2021 at 7:43 AM Adam Ford <aford...@gmail.com> wrote: > > On Fri, Mar 12, 2021 at 4:35 AM Marek Behún <marek.be...@nic.cz> wrote: > > > > Adam Ford says that DM3730 needs board.c compiled without LTO flags. > > > > Also add clock.c, since it says in Makefile that it need different > > flags. > > > > Tested-by: Adam Ford <aford...@gmail.com> #omap3_logic > > > Signed-off-by: Marek Behún <marek.be...@nic.cz> > > Suggested-by: Adam Ford <aford...@gmail.com> > > --- > > arch/arm/mach-omap2/omap3/Makefile | 2 ++ > > 1 file changed, 2 insertions(+) > > > > diff --git a/arch/arm/mach-omap2/omap3/Makefile > > b/arch/arm/mach-omap2/omap3/Makefile > > index 91ed8ebc9f..7d22f04bbf 100644 > > --- a/arch/arm/mach-omap2/omap3/Makefile > > +++ b/arch/arm/mach-omap2/omap3/Makefile > > @@ -5,10 +5,12 @@ > > > > # If clock.c is compiled for Thumb2, then it fails on OMAP3530 > > CFLAGS_clock.o += -marm > > +CFLAGS_REMOVE_clock.o := $(LTO_CFLAGS) > > I have two OMAP3530 boards. I'll test that this weekend. I think I > even have them configured for Thumb2
I have tested this on omap35_logic_somlv and the LTO flag removal isn't necessary for the clock.o Having the clock built with LTO saves 239 bytes in SPL with my compiler. It's not a huge savings, but in SPL, we need as much as possible. adam > > adam > > > > > obj-y := lowlevel_init.o > > > > obj-y += board.o > > +CFLAGS_REMOVE_board.o := $(LTO_CFLAGS) > > obj-y += boot.o > > obj-y += clock.o > > obj-y += sys_info.o > > -- > > 2.26.2 > >