1) + ((imask & 1 << i) >> i);
>> Again, formatting. Plus, i >> 1 << 1 looks too ugly/unreadable,
>> if you mean i & ~1, write it like that, it is up to the compiler to emit
>> it like i >> 1 << 1 if that is the best implementation.
>&
](if VL > 128, also use Imm[7:0]) provide 4 select
> > + controls for each element of the destination. */
> > + unsigned j = i % 4;
> > + sel_idx = ((i & 2) >> 1) * elems
> > + + (i >> 2 <
On Tue, Dec 15, 2020 at 06:10:57PM +0800, Hongtao Liu via Gcc-patches wrote:
> --- a/gcc/config/i386/i386.c
> +++ b/gcc/config/i386/i386.c
> @@ -18187,21 +18187,67 @@ ix86_gimple_fold_builtin (gimple_stmt_iterator *gsi)
> }
>break;
>
> +case IX86_BUILTIN_SHUFPD512:
> +case
From 74596b08a91dafcb29441de59544dd857a090564 Mon Sep 17 00:00:00 2001
From: liuhongt
Date: Fri, 11 Dec 2020 19:02:43 +0800
Subject: [PATCH] [X86] Fold more shuffle builtins to VEC_PERM_EXPR.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
A follow-up
