Re: [PATCH] ArmPkg/ArmLib ARM: set .fpu to let Clang 7 assemble ArmV7Support.S

[Ard Biesheuvel <ard.biesheuvel@linaro.org>]

On Fri, 21 Dec 2018 at 18:58, Leif Lindholm <leif.lindholm@linaro.org> wrote:
>
> On Thu, Dec 20, 2018 at 08:16:53PM +0100, Ard Biesheuvel wrote:
> > Clang 7 complains about the vmsr instruction in ArmV7Support.S,
> > which is only available on cores that implement some flavour of
> > VFP. So set the .fpu to NEON like we do in some other places.
> >
> > Contributed-under: TianoCore Contribution Agreement 1.1
> > Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
> > ---
> >  ArmPkg/Library/ArmLib/Arm/ArmV7Support.S | 1 +
> >  1 file changed, 1 insertion(+)
> >
> > diff --git a/ArmPkg/Library/ArmLib/Arm/ArmV7Support.S b/ArmPkg/Library/ArmLib/Arm/ArmV7Support.S
> > index 281499b46cbc..1808962ee3e2 100644
> > --- a/ArmPkg/Library/ArmLib/Arm/ArmV7Support.S
> > +++ b/ArmPkg/Library/ArmLib/Arm/ArmV7Support.S
> > @@ -268,6 +268,7 @@ ASM_FUNC(ArmEnableVFP)
> >  #ifndef __clang__
> >    mcr     p10,#0x7,r0,c8,c0,#0
> >  #else
> > +  .fpu    neon
> >    vmsr    fpexc, r0
> >  #endif
>
> No objection from me. But I would point out that the special clang
> filtering here could possibly be dropped.
>
> I mean, theoretically someone could have an even older binutils, but
> Linaro GCC 4.8-2013.05 will happily assemble the clang side of this
> conditional.
>

Yeah, but that would be a separate patch, no?