On 21 June 2014 13:58, Paolo Bonzini <pbonz...@redhat.com> wrote:
> Now that CPSR.E is set correctly, prepare for when setend will be able
> to change it; bswap data in and out of strex manually by comparing
> SCTLR.B, CPSR.E and TARGET_WORDS_BIGENDIAN (we do not have the luxury
> of using TCGMemOps).
>
> Signed-off-by: Paolo Bonzini <pbonz...@redhat.com>
> --- a/target-arm/cpu.h
> +++ b/target-arm/cpu.h
> @@ -1192,6 +1192,34 @@ static inline bool bswap_code(bool sctlr_b)
> #endif
> }
>
> +#ifdef CONFIG_USER_ONLY
> +/* get_user and put_user respectivaly return and expect data according
"respectively"
> + * to TARGET_WORDS_BIGENDIAN, but ldrex/strex emulation needs to take
> + * into account CPSR.E. Similar to bwap_code, a XOR gives exactly the
"bswap_code"
> + * required result, we just throw CPSR.E into the mix too:
> + *
> + * TARGET_WORDS_BIGENDIAN SCTLR.B CPSR.E need swap?
> + * LE/LE no 0 0 no
> + * LE/BE no 0 1 yes
> + * BE8/LE yes 0 0 yes
> + * BE8/BE yes 0 1 no
> + * BE32/BE yes 1 0 no
> + * (BE32/LE) yes 1 1 yes
> + *
> + * Officially, BE32 with CPSR.E=1 has "unpredictable" results. We
> + * implement it as big-endian code, little-endian data.
> + */
> +static inline bool arm_cpu_bswap_data(CPUARMState *env)
> +{
> + return
> +#ifdef TARGET_WORDS_BIGENDIAN
> + 1 ^
> +#endif
> + arm_sctlr_b(env) ^
> + !!(env->uncached_cpsr & CPSR_E);
> +}
> +#endif
Otherwise
Reviewed-by: Peter Maydell <peter.mayd...@linaro.org>
-- PMM
