On Fri, Oct 11, 2019 at 11:31:02AM -0400, Richard Henderson wrote:
> On 10/10/19 2:44 PM, Dave Martin wrote:
> > #define PSR_IL_BIT (1 << 20)
> > -#define PSR_BTYPE_CALL (2 << PSR_BTYPE_SHIFT)
> > +
> > +/* Convenience names for the values of PSTATE.BTYPE */
> > +#define PSR_BTYPE_NONE (0b00 << PSR_BTYPE_SHIFT)
> > +#define PSR_BTYPE_JC (0b01 << PSR_BTYPE_SHIFT)
> > +#define PSR_BTYPE_C (0b10 << PSR_BTYPE_SHIFT)
> > +#define PSR_BTYPE_J (0b11 << PSR_BTYPE_SHIFT)
>
> It'd be nice to sort this patch earlier, so that ...
>
> > diff --git a/arch/arm64/kernel/signal.c b/arch/arm64/kernel/signal.c
> > index 4a3bd32..452ac5b 100644
> > --- a/arch/arm64/kernel/signal.c
> > +++ b/arch/arm64/kernel/signal.c
> > @@ -732,7 +732,7 @@ static void setup_return(struct pt_regs *regs, struct k_sigaction *ka,
> >
> > if (system_supports_bti()) {
> > regs->pstate &= ~PSR_BTYPE_MASK;
> > - regs->pstate |= PSR_BTYPE_CALL;
> > + regs->pstate |= PSR_BTYPE_C;
> > }
> >
> > if (ka->sa.sa_flags & SA_RESTORER)
>
> ... setup_return does not need to be adjusted a second time.
>
> I don't see any other conflicts vs patch 5.
Ack, looks like I mis-split this during rebase.
Will fix.
Cheers
---Dave
