Re: [PATCH] RFC: use C11 alignas instead of GCC attribute aligned

[Tyler Retzlaff <roretzla@linux.microsoft.com>]

ping.

Please review this thread if you have time, the main point of discussion
I would like to receive consensus on the following questions.

1. Should we continue to expand common alignments behind an __rte_macro

  i.e. what do we prefer to appear in code

  alignas(RTE_CACHE_LINE_MIN_SIZE)

  -- or --

  __rte_cache_aligned

One of the benefits of dropping the macro is it provides a clear visual
indicator that it is not placed in the same location or get applied
to types as is done with __attribute__((__aligned__(n))).

2. where should we place alignas(n) or __rte_macro (if we use a macro)

Should it be on the same line as the variable or field or on the
preceeding line?

  /* same line example struct */
  struct T {
      /* alignas(64) applies to field0 *not* struct T type declaration */
      alignas(64) void *field0;
      void *field1;

      ... other fields ...

      alignas(64) uint64_t field5;
      uint32_t field6;

      ... more fields ...

  };

  /* same line example array */
  alignas(64) static const uint32_t array[4] = { ... };

  -- or --

  /* preceeding line example struct */
  struct T {
      /* alignas(64) applies to field0 *not* struct T type declaration */
      alignas(64)
      void *field0;
      void *field1;

      ... other fields ...

      alignas(64)
      uint64_t field5;
      uint32_t field6;

      ... more fields ...

  };

  /* preceeding line example array */
  alignas(64)
  static const uint32_t array[4] = { ... };


I'll submit patches for lib/* once the discussion is concluded.

thanks folks

On Wed, Nov 15, 2023 at 09:39:56AM -0800, Tyler Retzlaff wrote:
> Now that we require a C11 conformant toolchain we are able to improve
> portability by further adoption of C11 features.
> 
> Adapt EAL to use C11 alignas replacing __rte_cache_aligned and
> __rte_aligned(a) that expand to __attribute__((__aligned__(a))).
> 
> Note: it appears that use of alignas has exposed a bug in
>       lib/eal/riscv/include/rte_vect.h where the alignment
>       specified was reduced to 8 for xmm_t.
> 
> Please comment, subject to the outcome I will submit further series for
> lib/*
> 
> Thanks
> 
> Tyler Retzlaff (1):
>   eal: use C11 alignas instead of GCC attribute aligned
> 
>  lib/eal/arm/include/rte_vect.h       | 4 +++-
>  lib/eal/common/malloc_elem.h         | 4 +++-
>  lib/eal/common/malloc_heap.h         | 4 +++-
>  lib/eal/common/rte_keepalive.c       | 4 +++-
>  lib/eal/common/rte_random.c          | 5 ++++-
>  lib/eal/common/rte_service.c         | 7 +++++--
>  lib/eal/include/generic/rte_atomic.h | 4 +++-
>  lib/eal/loongarch/include/rte_vect.h | 7 +++++--
>  lib/eal/ppc/include/rte_vect.h       | 5 ++++-
>  lib/eal/riscv/include/rte_vect.h     | 4 +++-
>  lib/eal/x86/include/rte_vect.h       | 4 +++-
>  lib/eal/x86/rte_power_intrinsics.c   | 8 ++++++--
>  12 files changed, 45 insertions(+), 15 deletions(-)
> 
> -- 
> 1.8.3.1