[PATCH v3 4/5] arch,locking/atomic: hexagon: add arch_cmpxchg[64]_local

Brian Cain bcain at quicinc.com
Wed Nov 22 08:55:35 PST 2023



> -----Original Message-----
> From: wuqiang.matt <wuqiang.matt at bytedance.com>
> Sent: Tuesday, November 21, 2023 8:24 AM
> To: ubizjak at gmail.com; mark.rutland at arm.com; vgupta at kernel.org; Brian
> Cain <bcain at quicinc.com>; jonas at southpole.se;
> stefan.kristiansson at saunalahti.fi; shorne at gmail.com; chris at zankel.net;
> jcmvbkbc at gmail.com; geert at linux-m68k.org; andi.shyti at linux.intel.com;
> mingo at kernel.org; palmer at rivosinc.com; andrzej.hajda at intel.com;
> arnd at arndb.de; peterz at infradead.org; mhiramat at kernel.org
> Cc: linux-arch at vger.kernel.org; linux-snps-arc at lists.infradead.org; linux-
> kernel at vger.kernel.org; linux-hexagon at vger.kernel.org; linux-
> openrisc at vger.kernel.org; linux-trace-kernel at vger.kernel.org;
> mattwu at 163.com; linux at roeck-us.net; wuqiang.matt
> <wuqiang.matt at bytedance.com>; kernel test robot <lkp at intel.com>
> Subject: [PATCH v3 4/5] arch,locking/atomic: hexagon: add
> arch_cmpxchg[64]_local
> 
> WARNING: This email originated from outside of Qualcomm. Please be wary of
> any links or attachments, and do not enable macros.
> 
> hexagonc hasn't arch_cmpxhg_local implemented, which causes
> building failures for any references of try_cmpxchg_local,
> reported by the kernel test robot.
> 
> This patch implements arch_cmpxchg[64]_local with the native
> cmpxchg variant if the corresponding data size is supported,
> otherwise generci_cmpxchg[64]_local is to be used.
> 
> Reported-by: kernel test robot <lkp at intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202310272207.tLPflya4-
> lkp at intel.com/
> 
> Signed-off-by: wuqiang.matt <wuqiang.matt at bytedance.com>
> Reviewed-by: Masami Hiramatsu (Google) <mhiramat at kernel.org>
> ---
>  arch/hexagon/include/asm/cmpxchg.h | 51 +++++++++++++++++++++++++++++-
>  1 file changed, 50 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/hexagon/include/asm/cmpxchg.h
> b/arch/hexagon/include/asm/cmpxchg.h
> index bf6cf5579cf4..302fa30f25aa 100644
> --- a/arch/hexagon/include/asm/cmpxchg.h
> +++ b/arch/hexagon/include/asm/cmpxchg.h
> @@ -8,6 +8,8 @@
>  #ifndef _ASM_CMPXCHG_H
>  #define _ASM_CMPXCHG_H
> 
> +#include <linux/build_bug.h>
> +
>  /*
>   * __arch_xchg - atomically exchange a register and a memory location
>   * @x: value to swap
> @@ -51,13 +53,15 @@ __arch_xchg(unsigned long x, volatile void *ptr, int
> size)
>   *  variable casting.
>   */
> 
> -#define arch_cmpxchg(ptr, old, new)                            \
> +#define __cmpxchg_32(ptr, old, new)                            \
>  ({                                                             \
>         __typeof__(ptr) __ptr = (ptr);                          \
>         __typeof__(*(ptr)) __old = (old);                       \
>         __typeof__(*(ptr)) __new = (new);                       \
>         __typeof__(*(ptr)) __oldval = 0;                        \
>                                                                 \
> +       BUILD_BUG_ON(sizeof(*(ptr)) != 4);                      \
> +                                                               \
>         asm volatile(                                           \
>                 "1:     %0 = memw_locked(%1);\n"                \
>                 "       { P0 = cmp.eq(%0,%2);\n"                \
> @@ -72,4 +76,49 @@ __arch_xchg(unsigned long x, volatile void *ptr, int size)
>         __oldval;                                               \
>  })
> 
> +#define __cmpxchg(ptr, old, val, size)                         \
> +({                                                             \
> +       __typeof__(*(ptr)) oldval;                              \
> +                                                               \
> +       switch (size) {                                         \
> +       case 4:                                                 \
> +               oldval = __cmpxchg_32(ptr, old, val);           \
> +               break;                                          \
> +       default:                                                \
> +               BUILD_BUG();                                    \
> +               oldval = val;                                   \
> +               break;                                          \
> +       }                                                       \
> +                                                               \
> +       oldval;                                                 \
> +})
> +
> +#define arch_cmpxchg(ptr, o, n)        __cmpxchg((ptr), (o), (n), sizeof(*(ptr)))
> +
> +/*
> + * always make arch_cmpxchg[64]_local available, native cmpxchg
> + * will be used if available, then generic_cmpxchg[64]_local
> + */
> +#include <asm-generic/cmpxchg-local.h>
> +
> +#define arch_cmpxchg_local(ptr, old, val)                      \
> +({                                                             \
> +       __typeof__(*(ptr)) __retval;                            \
> +       int __size = sizeof(*(ptr));                            \
> +                                                               \
> +       switch (__size) {                                       \
> +       case 4:                                                 \
> +               __retval = __cmpxchg_32(ptr, old, val);         \
> +               break;                                          \
> +       default:                                                \
> +               __retval = __generic_cmpxchg_local(ptr, old,    \
> +                                               val, __size);   \
> +               break;                                          \
> +       }                                                       \
> +                                                               \
> +       __retval;                                               \
> +})
> +
> +#define arch_cmpxchg64_local(ptr, o, n) __generic_cmpxchg64_local((ptr),
> (o), (n))
> +
>  #endif /* _ASM_CMPXCHG_H */
> --
> 2.40.1

Acked-by: Brian Cain <bcain at quicinc.com>



More information about the linux-snps-arc mailing list