On 24/09/2020 01:17, Andre Przywara wrote: > When the actual offset between link and runtime address is zero, there > is no need for patching up U-Boot early when running with > CONFIG_POSITION_INDEPENDENT.
That turns out to be not fully true. Some toolchains (all Linaro cross compilers?) don't handle this well, they keep the original locations in the binary uninitialised, and rely on the reldyn fixup table to patch in the actual values. Other compilers (GCC 9.2 vanilla, Ubuntu GCC 7.5.0, Arm website 9.2) fill in the addresses both into the binary and the fixup, so this patch works. It seems to be fixed by enabling CONFIG_STATIC_RELA? I see it's disabled for CONFIG_POSITION_INDEPENDENT, what was the reason behind that? Cheers, Andre. > > Skip the whole routine when the distance is 0. > > This helps when U-Boot is loaded into ROM, or in otherwise sensitive > memory locations. > > Signed-off-by: Andre Przywara <andre.przyw...@arm.com> > --- > arch/arm/cpu/armv8/start.S | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/arch/arm/cpu/armv8/start.S b/arch/arm/cpu/armv8/start.S > index 002698b501c..02b952bb328 100644 > --- a/arch/arm/cpu/armv8/start.S > +++ b/arch/arm/cpu/armv8/start.S > @@ -66,7 +66,8 @@ save_boot_params_ret: > pie_fixup: > adr x0, _start /* x0 <- Runtime value of _start */ > ldr x1, _TEXT_BASE /* x1 <- Linked value of _start */ > - sub x9, x0, x1 /* x9 <- Run-vs-link offset */ > + subs x9, x0, x1 /* x9 <- Run-vs-link offset */ > + beq pie_fixup_done > adr x2, __rel_dyn_start /* x2 <- Runtime &__rel_dyn_start */ > adr x3, __rel_dyn_end /* x3 <- Runtime &__rel_dyn_end */ > pie_fix_loop: >
