From patchwork Wed May 27 20:15:34 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Christophe Lyon X-Patchwork-Id: 49040 Return-Path: X-Original-To: linaro@patches.linaro.org Delivered-To: linaro@patches.linaro.org Received: from mail-wi0-f200.google.com (mail-wi0-f200.google.com [209.85.212.200]) by ip-10-151-82-157.ec2.internal (Postfix) with ESMTPS id EDDF9202A3 for ; Wed, 27 May 2015 20:19:38 +0000 (UTC) Received: by wizk4 with SMTP id k4sf8174596wiz.2 for ; Wed, 27 May 2015 13:19:38 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:delivered-to:mailing-list :precedence:list-id:list-unsubscribe:list-archive:list-post :list-help:sender:delivered-to:from:to:subject:date:message-id :in-reply-to:references:x-original-sender :x-original-authentication-results; bh=MN7hI/cGKQuN3hKdD1z3XaPQDnUp5M8yja8KvLJABeU=; b=Nf/MiIj2FFS2JroRPZfObNoFPZY8+RUzcMm6riHwXiWTI/vqLnuorYqqx5EhkDk81U sPW9o1CbMjRr9pCoLQ8xLsAbo9Xq8mh0xoe4TLAFksMKk7+Ypv2U82i+snjS/zGQrObO zNrLLbowU92XomP8e+wgMtJIZBDKo9fjdc+YaBCiMzmZXFU0b6fVhbriLCvxE/YyQVZs ea4kz/gcg15aYozXJpjYFAor3G/BRa0h4T+4zWBj+bGxlq0rtwrbeWsNaOuw94ZM32mC dO+IdK/WUqzb/m8+MhQK6etYvIggx1bX+AH1W2h6p04ivxSW74C1LBiaysJ3cHvP/rvx jR7Q== X-Gm-Message-State: ALoCoQkcl9Ce+LaplhrIXqkDdJ+2A3c6WDX4JJE56BdMwRczAaHu/W4EdWz/8oF7e+TYxfuWEH0q X-Received: by 10.112.162.228 with SMTP id yd4mr33596733lbb.8.1432757978222; Wed, 27 May 2015 13:19:38 -0700 (PDT) MIME-Version: 1.0 X-BeenThere: patchwork-forward@linaro.org Received: by 10.152.120.71 with SMTP id la7ls107357lab.25.gmail; Wed, 27 May 2015 13:19:38 -0700 (PDT) X-Received: by 10.152.9.66 with SMTP id x2mr29675319laa.36.1432757978038; Wed, 27 May 2015 13:19:38 -0700 (PDT) Received: from mail-lb0-x229.google.com (mail-lb0-x229.google.com. [2a00:1450:4010:c04::229]) by mx.google.com with ESMTPS id yk6si33137lbb.90.2015.05.27.13.19.37 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 27 May 2015 13:19:37 -0700 (PDT) Received-SPF: pass (google.com: domain of patch+caf_=patchwork-forward=linaro.org@linaro.org designates 2a00:1450:4010:c04::229 as permitted sender) client-ip=2a00:1450:4010:c04::229; Received: by lbcmx3 with SMTP id mx3so15143201lbc.1 for ; Wed, 27 May 2015 13:19:37 -0700 (PDT) X-Received: by 10.112.222.133 with SMTP id qm5mr29828879lbc.86.1432757977571; Wed, 27 May 2015 13:19:37 -0700 (PDT) X-Forwarded-To: patchwork-forward@linaro.org X-Forwarded-For: patch@linaro.org patchwork-forward@linaro.org Delivered-To: patch@linaro.org Received: by 10.112.108.230 with SMTP id hn6csp627909lbb; Wed, 27 May 2015 13:19:35 -0700 (PDT) X-Received: by 10.70.90.209 with SMTP id by17mr60912560pdb.2.1432757975039; Wed, 27 May 2015 13:19:35 -0700 (PDT) Received: from sourceware.org (server1.sourceware.org. [209.132.180.131]) by mx.google.com with ESMTPS id fw16si30414pdb.179.2015.05.27.13.19.34 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 27 May 2015 13:19:35 -0700 (PDT) Received-SPF: pass (google.com: domain of gcc-patches-return-399123-patch=linaro.org@gcc.gnu.org designates 209.132.180.131 as permitted sender) client-ip=209.132.180.131; Received: (qmail 44994 invoked by alias); 27 May 2015 20:17:02 -0000 Mailing-List: list patchwork-forward@linaro.org; contact patchwork-forward+owners@linaro.org Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: , List-Help: , Sender: gcc-patches-owner@gcc.gnu.org Delivered-To: mailing list gcc-patches@gcc.gnu.org Received: (qmail 39802 invoked by uid 89); 27 May 2015 20:16:26 -0000 X-Virus-Found: No X-Spam-SWARE-Status: No, score=-2.2 required=5.0 tests=AWL, BAYES_00, RCVD_IN_DNSWL_NONE, SPF_PASS autolearn=ham version=3.3.2 X-HELO: mail-wg0-f45.google.com Received: from mail-wg0-f45.google.com (HELO mail-wg0-f45.google.com) (74.125.82.45) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-GCM-SHA256 encrypted) ESMTPS; Wed, 27 May 2015 20:16:04 +0000 Received: by wgv5 with SMTP id 5so19275480wgv.1 for ; Wed, 27 May 2015 13:16:00 -0700 (PDT) X-Received: by 10.180.221.9 with SMTP id qa9mr16042128wic.52.1432757760744; Wed, 27 May 2015 13:16:00 -0700 (PDT) Received: from babel.clyon.hd.free.fr (vig38-2-82-225-222-175.fbx.proxad.net. [82.225.222.175]) by mx.google.com with ESMTPSA id u7sm76992wif.3.2015.05.27.13.15.59 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 27 May 2015 13:16:00 -0700 (PDT) From: Christophe Lyon To: gcc-patches@gcc.gnu.org Subject: [Patch ARM-AArch64/testsuite Neon intrinsics 07/20] Add vrshr_n tests. Date: Wed, 27 May 2015 22:15:34 +0200 Message-Id: <1432757747-4891-8-git-send-email-christophe.lyon@linaro.org> In-Reply-To: <1432757747-4891-1-git-send-email-christophe.lyon@linaro.org> References: <1432757747-4891-1-git-send-email-christophe.lyon@linaro.org> X-IsSubscribed: yes X-Original-Sender: christophe.lyon@linaro.org X-Original-Authentication-Results: mx.google.com; spf=pass (google.com: domain of patch+caf_=patchwork-forward=linaro.org@linaro.org designates 2a00:1450:4010:c04::229 as permitted sender) smtp.mail=patch+caf_=patchwork-forward=linaro.org@linaro.org; dkim=pass header.i=@gcc.gnu.org X-Google-Group-Id: 836684582541 diff --git a/gcc/testsuite/gcc.target/aarch64/advsimd-intrinsics/vrshr_n.c b/gcc/testsuite/gcc.target/aarch64/advsimd-intrinsics/vrshr_n.c new file mode 100644 index 0000000..6f9ef5a --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/advsimd-intrinsics/vrshr_n.c @@ -0,0 +1,504 @@ +#include +#include "arm-neon-ref.h" +#include "compute-ref-data.h" + +/* Expected results. */ +VECT_VAR_DECL(expected,int,8,8) [] = { 0xf8, 0xf9, 0xf9, 0xfa, + 0xfa, 0xfb, 0xfb, 0xfc }; +VECT_VAR_DECL(expected,int,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected,int,32,2) [] = { 0xfffffffc, 0xfffffffc }; +VECT_VAR_DECL(expected,int,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected,uint,8,8) [] = { 0x3c, 0x3c, 0x3d, 0x3d, + 0x3d, 0x3d, 0x3e, 0x3e }; +VECT_VAR_DECL(expected,uint,16,4) [] = { 0x1ffe, 0x1ffe, 0x1ffe, 0x1ffe }; +VECT_VAR_DECL(expected,uint,32,2) [] = { 0x8000000, 0x8000000 }; +VECT_VAR_DECL(expected,uint,64,1) [] = { 0x80000000 }; +VECT_VAR_DECL(expected,int,8,16) [] = { 0xf8, 0xf9, 0xf9, 0xfa, + 0xfa, 0xfb, 0xfb, 0xfc, + 0xfc, 0xfd, 0xfd, 0xfe, + 0xfe, 0xff, 0xff, 0x0 }; +VECT_VAR_DECL(expected,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected,int,32,4) [] = { 0xfffffffc, 0xfffffffc, + 0xfffffffd, 0xfffffffd }; +VECT_VAR_DECL(expected,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected,uint,8,16) [] = { 0x3c, 0x3c, 0x3d, 0x3d, + 0x3d, 0x3d, 0x3e, 0x3e, + 0x3e, 0x3e, 0x3f, 0x3f, + 0x3f, 0x3f, 0x40, 0x40 }; +VECT_VAR_DECL(expected,uint,16,8) [] = { 0x1ffe, 0x1ffe, 0x1ffe, 0x1ffe, + 0x1fff, 0x1fff, 0x1fff, 0x1fff }; +VECT_VAR_DECL(expected,uint,32,4) [] = { 0x8000000, 0x8000000, + 0x8000000, 0x8000000 }; +VECT_VAR_DECL(expected,uint,64,2) [] = { 0x80000000, 0x80000000 }; + +/* Expected results with maximum input and max shift amount. */ +VECT_VAR_DECL(expected_max_sh_max,int,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,int,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,int,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,int,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,uint,8,8) [] = { 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_max,uint,16,4) [] = { 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_max,uint,32,2) [] = { 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_max,uint,64,1) [] = { 0x1 }; +VECT_VAR_DECL(expected_max_sh_max,int,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_max,uint,8,16) [] = { 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_max,uint,16,8) [] = { 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_max,uint,32,4) [] = { 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_max,uint,64,2) [] = { 0x1, 0x1 }; + +/* Expected results with maximum input and shift by 1. */ +VECT_VAR_DECL(expected_max_sh_1,int,8,8) [] = { 0x40, 0x40, 0x40, 0x40, + 0x40, 0x40, 0x40, 0x40 }; +VECT_VAR_DECL(expected_max_sh_1,int,16,4) [] = { 0x4000, 0x4000, + 0x4000, 0x4000 }; +VECT_VAR_DECL(expected_max_sh_1,int,32,2) [] = { 0x40000000, 0x40000000 }; +VECT_VAR_DECL(expected_max_sh_1,int,64,1) [] = { 0x4000000000000000 }; +VECT_VAR_DECL(expected_max_sh_1,uint,8,8) [] = { 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80 }; +VECT_VAR_DECL(expected_max_sh_1,uint,16,4) [] = { 0x8000, 0x8000, + 0x8000, 0x8000 }; +VECT_VAR_DECL(expected_max_sh_1,uint,32,2) [] = { 0x80000000, 0x80000000 }; +VECT_VAR_DECL(expected_max_sh_1,uint,64,1) [] = { 0x8000000000000000 }; +VECT_VAR_DECL(expected_max_sh_1,int,8,16) [] = { 0x40, 0x40, 0x40, 0x40, + 0x40, 0x40, 0x40, 0x40, + 0x40, 0x40, 0x40, 0x40, + 0x40, 0x40, 0x40, 0x40 }; +VECT_VAR_DECL(expected_max_sh_1,int,16,8) [] = { 0x4000, 0x4000, + 0x4000, 0x4000, + 0x4000, 0x4000, + 0x4000, 0x4000 }; +VECT_VAR_DECL(expected_max_sh_1,int,32,4) [] = { 0x40000000, 0x40000000, + 0x40000000, 0x40000000 }; +VECT_VAR_DECL(expected_max_sh_1,int,64,2) [] = { 0x4000000000000000, + 0x4000000000000000 }; +VECT_VAR_DECL(expected_max_sh_1,uint,8,16) [] = { 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80 }; +VECT_VAR_DECL(expected_max_sh_1,uint,16,8) [] = { 0x8000, 0x8000, + 0x8000, 0x8000, + 0x8000, 0x8000, + 0x8000, 0x8000 }; +VECT_VAR_DECL(expected_max_sh_1,uint,32,4) [] = { 0x80000000, 0x80000000, + 0x80000000, 0x80000000 }; +VECT_VAR_DECL(expected_max_sh_1,uint,64,2) [] = { 0x8000000000000000, + 0x8000000000000000 }; + +/* Expected results with maximum input and shift by 3. */ +VECT_VAR_DECL(expected_max_sh_3,int,8,8) [] = { 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10 }; +VECT_VAR_DECL(expected_max_sh_3,int,16,4) [] = { 0x1000, 0x1000, + 0x1000, 0x1000 }; +VECT_VAR_DECL(expected_max_sh_3,int,32,2) [] = { 0x10000000, 0x10000000 }; +VECT_VAR_DECL(expected_max_sh_3,int,64,1) [] = { 0x1000000000000000 }; +VECT_VAR_DECL(expected_max_sh_3,uint,8,8) [] = { 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20 }; +VECT_VAR_DECL(expected_max_sh_3,uint,16,4) [] = { 0x2000, 0x2000, + 0x2000, 0x2000 }; +VECT_VAR_DECL(expected_max_sh_3,uint,32,2) [] = { 0x20000000, 0x20000000 }; +VECT_VAR_DECL(expected_max_sh_3,uint,64,1) [] = { 0x2000000000000000 }; +VECT_VAR_DECL(expected_max_sh_3,int,8,16) [] = { 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10 }; +VECT_VAR_DECL(expected_max_sh_3,int,16,8) [] = { 0x1000, 0x1000, + 0x1000, 0x1000, + 0x1000, 0x1000, + 0x1000, 0x1000 }; +VECT_VAR_DECL(expected_max_sh_3,int,32,4) [] = { 0x10000000, 0x10000000, + 0x10000000, 0x10000000 }; +VECT_VAR_DECL(expected_max_sh_3,int,64,2) [] = { 0x1000000000000000, + 0x1000000000000000 }; +VECT_VAR_DECL(expected_max_sh_3,uint,8,16) [] = { 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20 }; +VECT_VAR_DECL(expected_max_sh_3,uint,16,8) [] = { 0x2000, 0x2000, + 0x2000, 0x2000, + 0x2000, 0x2000, + 0x2000, 0x2000 }; +VECT_VAR_DECL(expected_max_sh_3,uint,32,4) [] = { 0x20000000, 0x20000000, + 0x20000000, 0x20000000 }; +VECT_VAR_DECL(expected_max_sh_3,uint,64,2) [] = { 0x2000000000000000, + 0x2000000000000000 }; + +/* Expected results with max negative input (for signed types, shift + by 1. */ +VECT_VAR_DECL(expected_max_neg_sh_1,int,8,8) [] = { 0xc0, 0xc0, 0xc0, 0xc0, + 0xc0, 0xc0, 0xc0, 0xc0 }; +VECT_VAR_DECL(expected_max_neg_sh_1,int,16,4) [] = { 0xc000, 0xc000, + 0xc000, 0xc000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,int,32,2) [] = { 0xc0000000, 0xc0000000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,int,64,1) [] = { 0xc000000000000000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,8,8) [] = { 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,16,4) [] = { 0x8000, 0x8000, + 0x8000, 0x8000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,32,2) [] = { 0x80000000, 0x80000000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,64,1) [] = { 0x8000000000000000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,int,8,16) [] = { 0xc0, 0xc0, 0xc0, 0xc0, + 0xc0, 0xc0, 0xc0, 0xc0, + 0xc0, 0xc0, 0xc0, 0xc0, + 0xc0, 0xc0, 0xc0, 0xc0 }; +VECT_VAR_DECL(expected_max_neg_sh_1,int,16,8) [] = { 0xc000, 0xc000, + 0xc000, 0xc000, + 0xc000, 0xc000, + 0xc000, 0xc000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,int,32,4) [] = { 0xc0000000, 0xc0000000, + 0xc0000000, 0xc0000000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,int,64,2) [] = { 0xc000000000000000, + 0xc000000000000000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,8,16) [] = { 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,16,8) [] = { 0x8000, 0x8000, + 0x8000, 0x8000, + 0x8000, 0x8000, + 0x8000, 0x8000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,32,4) [] = { 0x80000000, 0x80000000, + 0x80000000, 0x80000000 }; +VECT_VAR_DECL(expected_max_neg_sh_1,uint,64,2) [] = { 0x8000000000000000, + 0x8000000000000000 }; + +/* Expected results with max negative input (for signed types, shift + by 3. */ +VECT_VAR_DECL(expected_max_neg_sh_3,int,8,8) [] = { 0xf0, 0xf0, 0xf0, 0xf0, + 0xf0, 0xf0, 0xf0, 0xf0 }; +VECT_VAR_DECL(expected_max_neg_sh_3,int,16,4) [] = { 0xf000, 0xf000, + 0xf000, 0xf000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,int,32,2) [] = { 0xf0000000, 0xf0000000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,int,64,1) [] = { 0xf000000000000000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,8,8) [] = { 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,16,4) [] = { 0x2000, 0x2000, + 0x2000, 0x2000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,32,2) [] = { 0x20000000, 0x20000000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,64,1) [] = { 0x2000000000000000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,int,8,16) [] = { 0xf0, 0xf0, 0xf0, 0xf0, + 0xf0, 0xf0, 0xf0, 0xf0, + 0xf0, 0xf0, 0xf0, 0xf0, + 0xf0, 0xf0, 0xf0, 0xf0 }; +VECT_VAR_DECL(expected_max_neg_sh_3,int,16,8) [] = { 0xf000, 0xf000, + 0xf000, 0xf000, + 0xf000, 0xf000, + 0xf000, 0xf000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,int,32,4) [] = { 0xf0000000, 0xf0000000, + 0xf0000000, 0xf0000000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,int,64,2) [] = { 0xf000000000000000, + 0xf000000000000000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,8,16) [] = { 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,16,8) [] = { 0x2000, 0x2000, + 0x2000, 0x2000, + 0x2000, 0x2000, + 0x2000, 0x2000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,32,4) [] = { 0x20000000, 0x20000000, + 0x20000000, 0x20000000 }; +VECT_VAR_DECL(expected_max_neg_sh_3,uint,64,2) [] = { 0x2000000000000000, + 0x2000000000000000 }; + +#define TEST_MSG "VRSHR_N" +void exec_vrshr_n (void) +{ + /* Basic test: y=vrshr_n(x,v), then store the result. */ +#define TEST_VRSHR_N(Q, T1, T2, W, N, V) \ + VECT_VAR(vector_res, T1, W, N) = \ + vrshr##Q##_n_##T2##W(VECT_VAR(vector, T1, W, N), \ + V); \ + vst1##Q##_##T2##W(VECT_VAR(result, T1, W, N), VECT_VAR(vector_res, T1, W, N)) + + DECL_VARIABLE_ALL_VARIANTS(vector); + DECL_VARIABLE_ALL_VARIANTS(vector_res); + + clean_results (); + + /* Initialize input "vector" from "buffer". */ + TEST_MACRO_ALL_VARIANTS_2_5(VLOAD, vector, buffer); + + /* Choose shift amount arbitrarily. */ + TEST_VRSHR_N(, int, s, 8, 8, 1); + TEST_VRSHR_N(, int, s, 16, 4, 12); + TEST_VRSHR_N(, int, s, 32, 2, 2); + TEST_VRSHR_N(, int, s, 64, 1, 32); + TEST_VRSHR_N(, uint, u, 8, 8, 2); + TEST_VRSHR_N(, uint, u, 16, 4, 3); + TEST_VRSHR_N(, uint, u, 32, 2, 5); + TEST_VRSHR_N(, uint, u, 64, 1, 33); + + TEST_VRSHR_N(q, int, s, 8, 16, 1); + TEST_VRSHR_N(q, int, s, 16, 8, 12); + TEST_VRSHR_N(q, int, s, 32, 4, 2); + TEST_VRSHR_N(q, int, s, 64, 2, 32); + TEST_VRSHR_N(q, uint, u, 8, 16, 2); + TEST_VRSHR_N(q, uint, u, 16, 8, 3); + TEST_VRSHR_N(q, uint, u, 32, 4, 5); + TEST_VRSHR_N(q, uint, u, 64, 2, 33); + +#define CMT "" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected, CMT); + + + /* Use maximum positive input value. */ + VDUP(vector, , int, s, 8, 8, 0x7F); + VDUP(vector, , int, s, 16, 4, 0x7FFF); + VDUP(vector, , int, s, 32, 2, 0x7FFFFFFF); + VDUP(vector, , int, s, 64, 1, 0x7FFFFFFFFFFFFFFFLL); + VDUP(vector, , uint, u, 8, 8, 0xFF); + VDUP(vector, , uint, u, 16, 4, 0xFFFF); + VDUP(vector, , uint, u, 32, 2, 0xFFFFFFFF); + VDUP(vector, , uint, u, 64, 1, 0xFFFFFFFFFFFFFFFFULL); + VDUP(vector, q, int, s, 8, 16, 0x7F); + VDUP(vector, q, int, s, 16, 8, 0x7FFF); + VDUP(vector, q, int, s, 32, 4, 0x7FFFFFFF); + VDUP(vector, q, int, s, 64, 2, 0x7FFFFFFFFFFFFFFFLL); + VDUP(vector, q, uint, u, 8, 16, 0xFF); + VDUP(vector, q, uint, u, 16, 8, 0xFFFF); + VDUP(vector, q, uint, u, 32, 4, 0xFFFFFFFF); + VDUP(vector, q, uint, u, 64, 2, 0xFFFFFFFFFFFFFFFFULL); + + /* Use max shift amount, to exercise saturation. */ + TEST_VRSHR_N(, int, s, 8, 8, 8); + TEST_VRSHR_N(, int, s, 16, 4, 16); + TEST_VRSHR_N(, int, s, 32, 2, 32); + TEST_VRSHR_N(, int, s, 64, 1, 64); + TEST_VRSHR_N(, uint, u, 8, 8, 8); + TEST_VRSHR_N(, uint, u, 16, 4, 16); + TEST_VRSHR_N(, uint, u, 32, 2, 32); + TEST_VRSHR_N(, uint, u, 64, 1, 64); + TEST_VRSHR_N(q, int, s, 8, 16, 8); + TEST_VRSHR_N(q, int, s, 16, 8, 16); + TEST_VRSHR_N(q, int, s, 32, 4, 32); + TEST_VRSHR_N(q, int, s, 64, 2, 64); + TEST_VRSHR_N(q, uint, u, 8, 16, 8); + TEST_VRSHR_N(q, uint, u, 16, 8, 16); + TEST_VRSHR_N(q, uint, u, 32, 4, 32); + TEST_VRSHR_N(q, uint, u, 64, 2, 64); + +#undef CMT +#define CMT " (overflow test: max shift amount, max positive input)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_max, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_max, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_max, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_max, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_max, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_max, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_max, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_max, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_max, CMT); + + + /* Use 1 as shift amount, to exercise saturation. */ + TEST_VRSHR_N(, int, s, 8, 8, 1); + TEST_VRSHR_N(, int, s, 16, 4, 1); + TEST_VRSHR_N(, int, s, 32, 2, 1); + TEST_VRSHR_N(, int, s, 64, 1, 1); + TEST_VRSHR_N(, uint, u, 8, 8, 1); + TEST_VRSHR_N(, uint, u, 16, 4, 1); + TEST_VRSHR_N(, uint, u, 32, 2, 1); + TEST_VRSHR_N(, uint, u, 64, 1, 1); + TEST_VRSHR_N(q, int, s, 8, 16, 1); + TEST_VRSHR_N(q, int, s, 16, 8, 1); + TEST_VRSHR_N(q, int, s, 32, 4, 1); + TEST_VRSHR_N(q, int, s, 64, 2, 1); + TEST_VRSHR_N(q, uint, u, 8, 16, 1); + TEST_VRSHR_N(q, uint, u, 16, 8, 1); + TEST_VRSHR_N(q, uint, u, 32, 4, 1); + TEST_VRSHR_N(q, uint, u, 64, 2, 1); + +#undef CMT +#define CMT " (overflow test: shift by 1, with max input)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_1, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_1, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_1, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_1, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_1, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_1, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_1, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_1, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_1, CMT); + + + /* Use 3 as shift amount, to exercise saturation. */ + TEST_VRSHR_N(, int, s, 8, 8, 3); + TEST_VRSHR_N(, int, s, 16, 4, 3); + TEST_VRSHR_N(, int, s, 32, 2, 3); + TEST_VRSHR_N(, int, s, 64, 1, 3); + TEST_VRSHR_N(, uint, u, 8, 8, 3); + TEST_VRSHR_N(, uint, u, 16, 4, 3); + TEST_VRSHR_N(, uint, u, 32, 2, 3); + TEST_VRSHR_N(, uint, u, 64, 1, 3); + TEST_VRSHR_N(q, int, s, 8, 16, 3); + TEST_VRSHR_N(q, int, s, 16, 8, 3); + TEST_VRSHR_N(q, int, s, 32, 4, 3); + TEST_VRSHR_N(q, int, s, 64, 2, 3); + TEST_VRSHR_N(q, uint, u, 8, 16, 3); + TEST_VRSHR_N(q, uint, u, 16, 8, 3); + TEST_VRSHR_N(q, uint, u, 32, 4, 3); + TEST_VRSHR_N(q, uint, u, 64, 2, 3); + +#undef CMT +#define CMT " (overflow test: shift by 3, with max input)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_3, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_3, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_3, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_3, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_3, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_3, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_3, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_3, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_3, CMT); + + + /* Use minimum negative input for signed types. */ + VDUP(vector, , int, s, 8, 8, 0x80); + VDUP(vector, , int, s, 16, 4, 0x8000); + VDUP(vector, , int, s, 32, 2, 0x80000000); + VDUP(vector, , int, s, 64, 1, 0x8000000000000000LL); + VDUP(vector, , uint, u, 8, 8, 0xFF); + VDUP(vector, , uint, u, 16, 4, 0xFFFF); + VDUP(vector, , uint, u, 32, 2, 0xFFFFFFFF); + VDUP(vector, , uint, u, 64, 1, 0xFFFFFFFFFFFFFFFFULL); + VDUP(vector, q, int, s, 8, 16, 0x80); + VDUP(vector, q, int, s, 16, 8, 0x8000); + VDUP(vector, q, int, s, 32, 4, 0x80000000); + VDUP(vector, q, int, s, 64, 2, 0x8000000000000000LL); + VDUP(vector, q, uint, u, 8, 16, 0xFF); + VDUP(vector, q, uint, u, 16, 8, 0xFFFF); + VDUP(vector, q, uint, u, 32, 4, 0xFFFFFFFF); + VDUP(vector, q, uint, u, 64, 2, 0xFFFFFFFFFFFFFFFFULL); + + + /* Use 1 as shift amount, to exercise saturation code. */ + TEST_VRSHR_N(, int, s, 8, 8, 1); + TEST_VRSHR_N(, int, s, 16, 4, 1); + TEST_VRSHR_N(, int, s, 32, 2, 1); + TEST_VRSHR_N(, int, s, 64, 1, 1); + TEST_VRSHR_N(, uint, u, 8, 8, 1); + TEST_VRSHR_N(, uint, u, 16, 4, 1); + TEST_VRSHR_N(, uint, u, 32, 2, 1); + TEST_VRSHR_N(, uint, u, 64, 1, 1); + TEST_VRSHR_N(q, int, s, 8, 16, 1); + TEST_VRSHR_N(q, int, s, 16, 8, 1); + TEST_VRSHR_N(q, int, s, 32, 4, 1); + TEST_VRSHR_N(q, int, s, 64, 2, 1); + TEST_VRSHR_N(q, uint, u, 8, 16, 1); + TEST_VRSHR_N(q, uint, u, 16, 8, 1); + TEST_VRSHR_N(q, uint, u, 32, 4, 1); + TEST_VRSHR_N(q, uint, u, 64, 2, 1); + +#undef CMT +#define CMT " (overflow test: shift by 1, with negative input)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_neg_sh_1, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_neg_sh_1, CMT); + + + /* Use 3 as shift amount, to exercise saturation code. */ + TEST_VRSHR_N(, int, s, 8, 8, 3); + TEST_VRSHR_N(, int, s, 16, 4, 3); + TEST_VRSHR_N(, int, s, 32, 2, 3); + TEST_VRSHR_N(, int, s, 64, 1, 3); + TEST_VRSHR_N(, uint, u, 8, 8, 3); + TEST_VRSHR_N(, uint, u, 16, 4, 3); + TEST_VRSHR_N(, uint, u, 32, 2, 3); + TEST_VRSHR_N(, uint, u, 64, 1, 3); + TEST_VRSHR_N(q, int, s, 8, 16, 3); + TEST_VRSHR_N(q, int, s, 16, 8, 3); + TEST_VRSHR_N(q, int, s, 32, 4, 3); + TEST_VRSHR_N(q, int, s, 64, 2, 3); + TEST_VRSHR_N(q, uint, u, 8, 16, 3); + TEST_VRSHR_N(q, uint, u, 16, 8, 3); + TEST_VRSHR_N(q, uint, u, 32, 4, 3); + TEST_VRSHR_N(q, uint, u, 64, 2, 3); + +#undef CMT +#define CMT " (overflow test: shift by 3, with negative input)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_neg_sh_3, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_neg_sh_3, CMT); +} + +int main (void) +{ + exec_vrshr_n (); + return 0; +}