[PATCH REPOST v2] ARM: virt: fix wrong HSCTLR.EE bit setting

[Marc Zyngier]

On 01/07/14 11:01, john.liuli wrote:
> From: Li Liu <john.liuli at huawei.com>
> 
> HSCTLR.EE is defined as bit[25] referring to arm manual
> DDI0606C.b(p1590).
> 
> Reviewed-by: Marc Zyngier <marc.zyngier at arm.com>
> Signed-off-by: Li Liu <john.liuli at huawei.com>
> ---
> changes v1 -> v2:
>   1)Use the canonical syntax for orr (orr r7, r7, #(1 << 25))
>   2)Use the ARM_BE8 macro instead of the open-coded CONFIG_CPU_BIG_ENDIAN
> 
>  arch/arm/kernel/hyp-stub.S |    4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S
> index 797b1a6..7e666cf 100644
> --- a/arch/arm/kernel/hyp-stub.S
> +++ b/arch/arm/kernel/hyp-stub.S
> @@ -134,9 +134,7 @@ ENTRY(__hyp_stub_install_secondary)
>  	mcr	p15, 4, r7, c1, c1, 3	@ HSTR
>  
>  THUMB(	orr	r7, #(1 << 30)	)	@ HSCTLR.TE
> -#ifdef CONFIG_CPU_BIG_ENDIAN
> -	orr	r7, #(1 << 9)		@ HSCTLR.EE
> -#endif
> +ARM_BE8(orr	r7, r7, #(1 << 25))     @ HSCTLR.EE
>  	mcr	p15, 4, r7, c1, c0, 0	@ HSCTLR
>  
>  	mrc	p15, 4, r7, c1, c1, 1	@ HDCR
> 

Russell,

Assuming you're happy with this patch, would you agree to it going
through the KVM/ARM tree? We have a whole bunch of BE patches queued up,
and this is a looks like a better fix than what we have so far [1].

Thanks,

	M.

[1]: https://lists.cs.columbia.edu/pipermail/kvmarm/2014-June/009950.html
-- 
Jazz is not dead. It just smells funny...