This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH] [powerpc] Use faster means to access FPSCR when possible in some cases
"Paul A. Clarke" <pc@us.ibm.com> writes:
> diff --git a/sysdeps/powerpc/fpu_control.h b/sysdeps/powerpc/fpu_control.h
> index fa04a67..240eae7 100644
> --- a/sysdeps/powerpc/fpu_control.h
> +++ b/sysdeps/powerpc/fpu_control.h
> @@ -65,35 +65,34 @@ extern fpu_control_t __fpu_control;
>...
> -#ifdef _ARCH_PWR9
> -# define __FPU_MFFSL() \
> - ({register double __fr; \
> - __asm__ __volatile__("mffsl %0" : "=f" (__fr)); \
> - __fr; \
> +# define _FPU_GET_RC_ISA300() \
> + ({union { double __d; unsigned long long __ll; } __u; \
> + __asm__ __volatile__( \
> + ".machine push; .machine \"power9\"; mffsl %0; .machine pop" \
> + : "=f" (__u.__d)); \
> + (fpu_control_t) (__u.__ll & _FPU_MASK_RC); \
> })
> +
> +#ifdef _ARCH_PWR9
> +# define _FPU_GET_RC() _FPU_GET_RC_ISA300()
> +#elif defined __BUILTIN_CPU_SUPPORTS__
> +# define _FPU_GET_RC() \
> + ({fpu_control_t __rc; \
> + __rc = __glibc_likely (__builtin_cpu_supports ("arch_3_00")) \
> + ? _FPU_GET_RC_ISA300 () \
> + : _FPU_GETCW (__rc); \
Missing the mask _FPU_MASK_RC.
Fixed.
> + __rc; \
> + })
> #else
> -# define __FPU_MFFSL() __FPU_MFFS()
> +# define _FPU_GET_RC() ({ fpu_control_t __rc = _FPU_GETCW (__rc); __rc; })
Likewise. Fixed.
> #endif
> -
> -# define _FPU_GET_RC() \
> - ({union { double __d; unsigned long long __ll; } __u; \
> - __u.__d = __FPU_MFFSL(); \
> - __u.__ll &= _FPU_MASK_RC; \
> - (fpu_control_t) __u.__ll; \
> - })
I'm merging the patch with those fixes.
Thanks!
--
Tulio Magno