This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: [PATCH] x86: fold AVX scalar to/from int conversion insns
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Jan Beulich <JBeulich at suse dot com>
- Cc: Binutils <binutils at sourceware dot org>
- Date: Mon, 24 Jun 2019 09:31:14 -0700
- Subject: Re: [PATCH] x86: fold AVX scalar to/from int conversion insns
- References: <5D10D9B7020000780023AAB4@prv1-mh.provo.novell.com>
On Mon, Jun 24, 2019 at 7:10 AM Jan Beulich <JBeulich@suse.com> wrote:
>
> There's no point doing a separate decode of the VEX.L bit - both decoded
> forms are identical.
>
> opcodes/
> 2019-06-24 Jan Beulich <jbeulich@suse.com>
>
> * i386-dis.c (VEX_LEN_0F2A_P_1, VEX_LEN_0F2A_P_3,
> VEX_LEN_0F2C_P_1, VEX_LEN_0F2C_P_3, VEX_LEN_0F2D_P_1,
> VEX_LEN_0F2D_P_3): Delete.
> (vex_len_table): Move vcvtsi2ss, vcvtsi2sd, vcvttss2si,
> vcvttsd2si, vcvtss2si, and vcvtsd2si leaf entries ...
> (prefix_table): ... here.
>
OK.
Thanks.
--
H.J.