On Thu, Jun 03, 2021 at 04:40:24PM +0100, Dave Martin wrote: > On Fri, May 21, 2021 at 03:46:20PM +0100, Mark Brown wrote: > > - if (system_supports_bti() && > > + if (system_supports_bti() && is_interp && > Won't this cause BTI to be forced off for static binaries? > Perhaps this should be (has_interp == is_interp), as for > arch_elf_adjust_prot(). Seems gross though, since has_interp would > become useless after the next patch. If there's no sensible way to > keep this bisectable, perhaps the patches can be merged instead. Ugh, right. I only tested the finished result.
Attachment:
signature.asc
Description: PGP signature
