On Mon, Mar 23, 2020 at 02:39:55PM +0000, Catalin Marinas wrote: > On Mon, Mar 23, 2020 at 01:57:22PM +0000, Mark Rutland wrote: > > On Mon, Mar 23, 2020 at 01:24:12PM +0000, Mark Brown wrote: > > > On Mon, Mar 23, 2020 at 12:21:44PM +0000, Catalin Marinas wrote: > > > > On Fri, Mar 20, 2020 at 05:39:46PM +0000, Szabolcs Nagy wrote: > > > > > > > +int arch_elf_adjust_prot(int prot, const struct arch_elf_state *state, > > > > + bool has_interp, bool is_interp) > > > > +{ > > > > + if (is_interp != has_interp) > > > > + return prot; > > > > + > > > > + if (!(state->flags & ARM64_ELF_BTI)) > > > > + return prot; > > > > + > > > > + if (prot & PROT_EXEC) > > > > + prot |= PROT_BTI; > > > > + > > > > + return prot; > > > > +} > > I think it would be best to document the current behaviour, as it's a > > simple ABI that we can guarantee, and the dynamic linker will have to be > > aware of BTI in order to do the right thing anyhow. > > That's a valid point. If we have an old dynamic linker and the kernel > enabled BTI automatically for the main executable, could things go wrong > (e.g. does the PLT need to be BTI-aware)? Also worth noting that an old dynamic linker won't have ARM64_ELF_BTI set, so the kernel will not enable BTI for this. Mark.