[i386,AVX512,21/n] Extend variable shift patterns.

Message ID	20140822101453.GA47539@msticlxl57.ims.intel.com
State	New
Headers	show Return-Path: <gcc-patches-return-375964-incoming=patchwork.ozlabs.org@gcc.gnu.org> DomainKey-Signature: a=rsa-sha1; c=nofws; d=gcc.gnu.org; h=list-id :list-unsubscribe:list-archive:list-post:list-help:sender:date :from:to:cc:subject:message-id:mime-version:content-type; q=dns; s=default; b=fblWJ9a+O4x0ZYuuLE6KOGmuJmEgIkI+u8/WATizAyuHPXfUBA DFZLvEPSKO/NATpfHTlNdQXxukpdN9QHsShivpDj5L/ye45Jj0xer1JCleZISOwI p9BpcJV1KdoiwsZXeyEbuqGEAJ8d0dABPzJ4zDZaq75A8Tso93V5orW+U= Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk Sender: gcc-patches-owner@gcc.gnu.org Date: Fri, 22 Aug 2014 14:15:01 +0400 From: Kirill Yukhin <kirill.yukhin@gmail.com> To: Uros Bizjak <ubizjak@gmail.com> Cc: Jakub Jelinek <jakub@redhat.com>, Richard Henderson <rth@redhat.com>, GCC Patches <gcc-patches@gcc.gnu.org>, kirill.yukhin@gmail.com Subject: [PATCH i386 AVX512] [21/n] Extend variable shift patterns. Message-ID: <20140822101453.GA47539@msticlxl57.ims.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.21 (2010-09-15)

Message ID

20140822101453.GA47539@msticlxl57.ims.intel.com

State

New

Headers

DomainKey-Signature: a=rsa-sha1; c=nofws; d=gcc.gnu.org; h=list-id
	:list-unsubscribe:list-archive:list-post:list-help:sender:date
	:from:to:cc:subject:message-id:mime-version:content-type; q=dns;
	s=default; b=fblWJ9a+O4x0ZYuuLE6KOGmuJmEgIkI+u8/WATizAyuHPXfUBA
	DFZLvEPSKO/NATpfHTlNdQXxukpdN9QHsShivpDj5L/ye45Jj0xer1JCleZISOwI
	p9BpcJV1KdoiwsZXeyEbuqGEAJ8d0dABPzJ4zDZaq75A8Tso93V5orW+U=
Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
Sender: gcc-patches-owner@gcc.gnu.org
Date: Fri, 22 Aug 2014 14:15:01 +0400
From: Kirill Yukhin <kirill.yukhin@gmail.com>
To: Uros Bizjak <ubizjak@gmail.com>
Cc: Jakub Jelinek <jakub@redhat.com>, Richard Henderson <rth@redhat.com>,
	GCC Patches <gcc-patches@gcc.gnu.org>, kirill.yukhin@gmail.com
Subject: [PATCH i386 AVX512] [21/n] Extend variable shift patterns.
Message-ID: <20140822101453.GA47539@msticlxl57.ims.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.21 (2010-09-15)

Commit Message

Kirill Yukhin Aug. 22, 2014, 10:15 a.m. UTC

Hello,
This patch extends shift patterns with per-element
shift value. It was adopted by approach suggested
in previous patches.

Bootstrapped.
New tests on top of patch-set all pass
under simulator.

Is it ok for trunk?

gcc/
	* config/i386/sse.md
	(define_mode_iterator VI48_AVX2_48_AVX512F): Delete.
	(define_mode_iterator VI48_AVX512BW): New.
	(define_insn "<avx2_avx512f>_<shift_insn>v<mode><mask_name>"): Delete.
	(define_insn "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>"
	with VI48_AVX2_48_AVX512F): New.
	(define_insn "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>"
	with VI2_AVX512VL): Ditto.

--
Thanks, K

Comments

Uros Bizjak Aug. 23, 2014, 7:27 a.m. UTC | #1

On Fri, Aug 22, 2014 at 12:15 PM, Kirill Yukhin <kirill.yukhin@gmail.com> wrote:

> This patch extends shift patterns with per-element
> shift value. It was adopted by approach suggested
> in previous patches.
>
> Bootstrapped.
> New tests on top of patch-set all pass
> under simulator.
>
> Is it ok for trunk?
>
> gcc/
>         * config/i386/sse.md
>         (define_mode_iterator VI48_AVX2_48_AVX512F): Delete.
>         (define_mode_iterator VI48_AVX512BW): New.
>         (define_insn "<avx2_avx512f>_<shift_insn>v<mode><mask_name>"): Delete.
>         (define_insn "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>"
>         with VI48_AVX2_48_AVX512F): New.
>         (define_insn "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>"
>         with VI2_AVX512VL): Ditto.

OK.

Thanks,
Uros.

diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 5084892..cd0c08e 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -359,9 +359,9 @@ 
    (V8SI "TARGET_AVX2") V4SI
    (V8DI "TARGET_AVX512F") (V4DI "TARGET_AVX2") V2DI])
 
-(define_mode_iterator VI48_AVX2_48_AVX512F
-  [(V16SI "TARGET_AVX512F") (V8SI "TARGET_AVX2") V4SI
-   (V8DI "TARGET_AVX512F") (V4DI "TARGET_AVX2") V2DI])
+(define_mode_iterator VI48_AVX512F
+  [(V16SI "TARGET_AVX512F") V8SI V4SI
+   (V8DI "TARGET_AVX512F") V4DI V2DI])
 
 (define_mode_iterator V48_AVX2
   [V4SF V2DF
@@ -15320,17 +15320,28 @@ 
    (set_attr "prefix" "maybe_evex")
    (set_attr "mode" "<sseinsnmode>")])
 
-(define_insn "<avx2_avx512f>_<shift_insn>v<mode><mask_name>"
-  [(set (match_operand:VI48_AVX2_48_AVX512F 0 "register_operand" "=v")
-	(any_lshift:VI48_AVX2_48_AVX512F
-	  (match_operand:VI48_AVX2_48_AVX512F 1 "register_operand" "v")
-	  (match_operand:VI48_AVX2_48_AVX512F 2 "nonimmediate_operand" "vm")))]
+(define_insn "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>"
+  [(set (match_operand:VI48_AVX512F 0 "register_operand" "=v")
+	(any_lshift:VI48_AVX512F
+	  (match_operand:VI48_AVX512F 1 "register_operand" "v")
+	  (match_operand:VI48_AVX512F 2 "nonimmediate_operand" "vm")))]
   "TARGET_AVX2 && <mask_mode512bit_condition>"
   "vp<vshift>v<ssemodesuffix>\t{%2, %1, %0<mask_operand3>|%0<mask_operand3>, %1, %2}"
   [(set_attr "type" "sseishft")
    (set_attr "prefix" "maybe_evex")
    (set_attr "mode" "<sseinsnmode>")])
 
+(define_insn "<avx2_avx512bw>_<shift_insn>v<mode><mask_name>"
+  [(set (match_operand:VI2_AVX512VL 0 "register_operand" "=v")
+	(any_lshift:VI2_AVX512VL
+	  (match_operand:VI2_AVX512VL 1 "register_operand" "v")
+	  (match_operand:VI2_AVX512VL 2 "nonimmediate_operand" "vm")))]
+  "TARGET_AVX512BW"
+  "vp<vshift>v<ssemodesuffix>\t{%2, %1, %0<mask_operand3>|%0<mask_operand3>, %1, %2}"
+  [(set_attr "type" "sseishft")
+   (set_attr "prefix" "maybe_evex")
+   (set_attr "mode" "<sseinsnmode>")])
+
 ;; For avx_vec_concat<mode> insn pattern
 (define_mode_attr concat_tg_mode
   [(V32QI "t") (V16HI "t") (V8SI "t") (V4DI "t") (V8SF "t") (V4DF "t")