On Thu, Apr 10, 2025 at 4:09 PM Richard Biener <rguent...@suse.de> wrote:
>
> The following removes the ability to switch back to non SLP-only
> operation of the vectorizer - a requirement to start cleaning out
> non-SLP paths without risk of regressing that case.

I have pushed this series now.

Richard.

>         * params.opt (--param=vect-force-slp): Remove.
>         * doc/invoke.texi (--param=vect-force-slp): Likewise.
>         * tree-vect-loop.cc (vect_analyze_loop_2): Assume
>         param_vect_force_slp is 1.
>         * tree-vect-stmts.cc (vect_analyze_stmt): Likewise.
> ---
>  gcc/doc/invoke.texi    | 3 ---
>  gcc/params.opt         | 4 ----
>  gcc/tree-vect-loop.cc  | 2 +-
>  gcc/tree-vect-stmts.cc | 2 +-
>  4 files changed, 2 insertions(+), 9 deletions(-)
>
> diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
> index a040d65af0e..7ec2498a12d 100644
> --- a/gcc/doc/invoke.texi
> +++ b/gcc/doc/invoke.texi
> @@ -17171,9 +17171,6 @@ this parameter.  The default value of this parameter 
> is 50.
>  @item vect-induction-float
>  Enable loop vectorization of floating point inductions.
>
> -@item vect-force-slp
> -Force the use of SLP when vectorizing, fail if not possible.
> -
>  @item vrp-block-limit
>  Maximum number of basic blocks before VRP switches to a lower memory 
> algorithm.
>
> diff --git a/gcc/params.opt b/gcc/params.opt
> index 422d082b355..748272c9fb9 100644
> --- a/gcc/params.opt
> +++ b/gcc/params.opt
> @@ -1230,10 +1230,6 @@ The maximum factor which the loop vectorizer applies 
> to the cost of statements i
>  Common Joined UInteger Var(param_vect_induction_float) Init(1) 
> IntegerRange(0, 1) Param Optimization
>  Enable loop vectorization of floating point inductions.
>
> --param=vect-force-slp=
> -Common Joined UInteger Var(param_vect_force_slp) Init(1) IntegerRange(0, 1) 
> Param Optimization
> -Force the use of SLP when vectorizing, fail if not possible.
> -
>  -param=vrp-block-limit=
>  Common Joined UInteger Var(param_vrp_block_limit) Init(150000) Optimization 
> Param
>  Maximum number of basic blocks before VRP switches to a fast model with less 
> memory requirements.
> diff --git a/gcc/tree-vect-loop.cc b/gcc/tree-vect-loop.cc
> index 9413dcef702..792231e5cf6 100644
> --- a/gcc/tree-vect-loop.cc
> +++ b/gcc/tree-vect-loop.cc
> @@ -2928,7 +2928,7 @@ start_over:
>
>    /* When we arrive here with SLP disabled and we are supposed
>       to use SLP for everything fail vectorization.  */
> -  if (!slp && param_vect_force_slp)
> +  if (!slp)
>      return opt_result::failure_at (vect_location,
>                                    "may need non-SLP handling\n");
>
> diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc
> index 7f874354e75..5a1ab7e0e26 100644
> --- a/gcc/tree-vect-stmts.cc
> +++ b/gcc/tree-vect-stmts.cc
> @@ -14006,7 +14006,7 @@ vect_analyze_stmt (vec_info *vinfo,
>
>    /* When we arrive here with a non-SLP statement and we are supposed
>       to use SLP for everything fail vectorization.  */
> -  if (!node && param_vect_force_slp)
> +  if (!node)
>      return opt_result::failure_at (stmt_info->stmt,
>                                    "needs non-SLP handling\n");
>
> --
> 2.43.0
>

Reply via email to