From patchwork Wed May 27 20:15:32 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Christophe Lyon X-Patchwork-Id: 49041 Return-Path: X-Original-To: linaro@patches.linaro.org Delivered-To: linaro@patches.linaro.org Received: from mail-la0-f70.google.com (mail-la0-f70.google.com [209.85.215.70]) by ip-10-151-82-157.ec2.internal (Postfix) with ESMTPS id CF9CE202A3 for ; Wed, 27 May 2015 20:19:52 +0000 (UTC) Received: by lani11 with SMTP id i11sf5604456lan.3 for ; Wed, 27 May 2015 13:19:51 -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=Q7CAXZnUo//nVfsLtEZ9Xp0xyHg6KOj5v6xIsaWoOS4=; b=VVSn/D2hzE7Fdn40pYUyzXAntHX5wybIgTQ4hfWEEqPdc9Os4oLnqqjzxFRjYFCc4z kSIPPV7EXU44SPrlLb1LHDpvwcJDJggrHrJdpJbTChyOWzQAwekj8toW6kSIJ2ivhkCo CAGJTU5M+KLATZQPuDs9lLrZq8+x0cccz6j9ixMSvQSXVe5hnUd9ivr7+jQX73V3nTZA 2QwNa39LNxbC5D2sfxGv077RANjpSka5n/ZIrJ1LF50bUzmXciw0/iEKcc6NjBmceSeU t7yzdmEn8gdLLjmG92Bg/S6w4l3EoWxNbjVGB6ZhgkXDEGyXHJuMQxCK+L4OXlHDCN1R gLvA== X-Gm-Message-State: ALoCoQlznNitcJ5nT0kBtm6JCa90MObanWP3xo6MWeYrQNmKSPPs8xp+CxJAEEU1t3xb7pJLb1R6 X-Received: by 10.112.29.39 with SMTP id g7mr33846042lbh.1.1432757991478; Wed, 27 May 2015 13:19:51 -0700 (PDT) MIME-Version: 1.0 X-BeenThere: patchwork-forward@linaro.org Received: by 10.152.203.133 with SMTP id kq5ls94966lac.102.gmail; Wed, 27 May 2015 13:19:51 -0700 (PDT) X-Received: by 10.152.5.98 with SMTP id r2mr16125038lar.8.1432757991316; Wed, 27 May 2015 13:19:51 -0700 (PDT) Received: from mail-la0-x22b.google.com (mail-la0-x22b.google.com. [2a00:1450:4010:c03::22b]) by mx.google.com with ESMTPS id yg5si9832lab.174.2015.05.27.13.19.51 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 27 May 2015 13:19:51 -0700 (PDT) Received-SPF: pass (google.com: domain of patch+caf_=patchwork-forward=linaro.org@linaro.org designates 2a00:1450:4010:c03::22b as permitted sender) client-ip=2a00:1450:4010:c03::22b; Received: by lagv1 with SMTP id v1so16138374lag.3 for ; Wed, 27 May 2015 13:19:51 -0700 (PDT) X-Received: by 10.112.93.37 with SMTP id cr5mr11595853lbb.106.1432757991214; Wed, 27 May 2015 13:19:51 -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 hn6csp628012lbb; Wed, 27 May 2015 13:19:49 -0700 (PDT) X-Received: by 10.70.138.70 with SMTP id qo6mr60732909pdb.163.1432757988387; Wed, 27 May 2015 13:19:48 -0700 (PDT) Received: from sourceware.org (server1.sourceware.org. [209.132.180.131]) by mx.google.com with ESMTPS id z2si86888pdj.39.2015.05.27.13.19.47 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Wed, 27 May 2015 13:19:48 -0700 (PDT) Received-SPF: pass (google.com: domain of gcc-patches-return-399124-patch=linaro.org@gcc.gnu.org designates 209.132.180.131 as permitted sender) client-ip=209.132.180.131; Received: (qmail 45197 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 40433 invoked by uid 89); 27 May 2015 20:16:29 -0000 X-Virus-Found: No X-Spam-SWARE-Status: No, score=-2.6 required=5.0 tests=AWL, BAYES_00, RCVD_IN_DNSWL_LOW, SPF_PASS autolearn=ham version=3.3.2 X-HELO: mail-wi0-f169.google.com Received: from mail-wi0-f169.google.com (HELO mail-wi0-f169.google.com) (209.85.212.169) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-GCM-SHA256 encrypted) ESMTPS; Wed, 27 May 2015 20:16:02 +0000 Received: by wizk4 with SMTP id k4so124255631wiz.1 for ; Wed, 27 May 2015 13:15:58 -0700 (PDT) X-Received: by 10.180.91.17 with SMTP id ca17mr15998857wib.9.1432757758880; Wed, 27 May 2015 13:15:58 -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.3.2015.05.27.13.15.57 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Wed, 27 May 2015 13:15:58 -0700 (PDT) From: Christophe Lyon To: gcc-patches@gcc.gnu.org Subject: [Patch ARM-AArch64/testsuite Neon intrinsics 05/20] Add vrshl tests. Date: Wed, 27 May 2015 22:15:32 +0200 Message-Id: <1432757747-4891-6-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:c03::22b 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/vrshl.c b/gcc/testsuite/gcc.target/aarch64/advsimd-intrinsics/vrshl.c new file mode 100644 index 0000000..d970fbd --- /dev/null +++ b/gcc/testsuite/gcc.target/aarch64/advsimd-intrinsics/vrshl.c @@ -0,0 +1,627 @@ +#include +#include "arm-neon-ref.h" +#include "compute-ref-data.h" + +/* Expected results with input=0. */ +VECT_VAR_DECL(expected_0,int,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,int,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,int,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,int,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_0,uint,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,uint,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,uint,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,uint,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_0,int,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,uint,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0,uint,64,2) [] = { 0x0, 0x0 }; + +/* Expected results with input=0 and negative shift amount. */ +VECT_VAR_DECL(expected_0_sh_neg,int,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,int,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,int,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,int,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,int,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_0_sh_neg,uint,64,2) [] = { 0x0, 0x0 }; + +/* Expected results. */ +VECT_VAR_DECL(expected,int,8,8) [] = { 0xe0, 0xe2, 0xe4, 0xe6, + 0xe8, 0xea, 0xec, 0xee }; +VECT_VAR_DECL(expected,int,16,4) [] = { 0xff80, 0xff88, 0xff90, 0xff98 }; +VECT_VAR_DECL(expected,int,32,2) [] = { 0xfffff000, 0xfffff100 }; +VECT_VAR_DECL(expected,int,64,1) [] = { 0xfffffffffffffffe }; +VECT_VAR_DECL(expected,uint,8,8) [] = { 0xe0, 0xe2, 0xe4, 0xe6, + 0xe8, 0xea, 0xec, 0xee }; +VECT_VAR_DECL(expected,uint,16,4) [] = { 0xff80, 0xff88, 0xff90, 0xff98 }; +VECT_VAR_DECL(expected,uint,32,2) [] = { 0xfffff000, 0xfffff100 }; +VECT_VAR_DECL(expected,uint,64,1) [] = { 0x1ffffffffffffffe }; +VECT_VAR_DECL(expected,int,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected,int,16,8) [] = { 0x0, 0x1000, 0x2000, 0x3000, + 0x4000, 0x5000, 0x6000, 0x7000 }; +VECT_VAR_DECL(expected,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected,int,64,2) [] = { 0x0, 0x8000000000000000 }; +VECT_VAR_DECL(expected,uint,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected,uint,16,8) [] = { 0x0, 0x1000, 0x2000, 0x3000, + 0x4000, 0x5000, 0x6000, 0x7000 }; +VECT_VAR_DECL(expected,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected,uint,64,2) [] = { 0x0, 0x8000000000000000 }; + +/* Expected results with negative shift amount. */ +VECT_VAR_DECL(expected_sh_neg,int,8,8) [] = { 0xf8, 0xf9, 0xf9, 0xfa, + 0xfa, 0xfb, 0xfb, 0xfc }; +VECT_VAR_DECL(expected_sh_neg,int,16,4) [] = { 0xfffc, 0xfffc, 0xfffd, 0xfffd }; +VECT_VAR_DECL(expected_sh_neg,int,32,2) [] = { 0xfffffffe, 0xfffffffe }; +VECT_VAR_DECL(expected_sh_neg,int,64,1) [] = { 0xffffffffffffffff }; +VECT_VAR_DECL(expected_sh_neg,uint,8,8) [] = { 0x78, 0x79, 0x79, 0x7a, + 0x7a, 0x7b, 0x7b, 0x7c }; +VECT_VAR_DECL(expected_sh_neg,uint,16,4) [] = { 0x3ffc, 0x3ffc, 0x3ffd, 0x3ffd }; +VECT_VAR_DECL(expected_sh_neg,uint,32,2) [] = { 0x1ffffffe, 0x1ffffffe }; +VECT_VAR_DECL(expected_sh_neg,uint,64,1) [] = { 0xfffffffffffffff }; +VECT_VAR_DECL(expected_sh_neg,int,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_sh_neg,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_sh_neg,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_sh_neg,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_sh_neg,uint,8,16) [] = { 0x2, 0x2, 0x2, 0x2, + 0x2, 0x2, 0x2, 0x2, + 0x2, 0x2, 0x2, 0x2, + 0x2, 0x2, 0x2, 0x2 }; +VECT_VAR_DECL(expected_sh_neg,uint,16,8) [] = { 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20 }; +VECT_VAR_DECL(expected_sh_neg,uint,32,4) [] = { 0x80000, 0x80000, + 0x80000, 0x80000 }; +VECT_VAR_DECL(expected_sh_neg,uint,64,2) [] = { 0x100000000000, 0x100000000000 }; + +/* Expected results with max input value shifted by -1 to test + round_const. */ +VECT_VAR_DECL(expected_max_sh_minus1,int,8,8) [] = { 0x40, 0x40, 0x40, 0x40, + 0x40, 0x40, 0x40, 0x40 }; +VECT_VAR_DECL(expected_max_sh_minus1,int,16,4) [] = { 0x4000, 0x4000, + 0x4000, 0x4000 }; +VECT_VAR_DECL(expected_max_sh_minus1,int,32,2) [] = { 0x40000000, 0x40000000 }; +VECT_VAR_DECL(expected_max_sh_minus1,int,64,1) [] = { 0x4000000000000000 }; +VECT_VAR_DECL(expected_max_sh_minus1,uint,8,8) [] = { 0x80, 0x80, 0x80, 0x80, + 0x80, 0x80, 0x80, 0x80 }; +VECT_VAR_DECL(expected_max_sh_minus1,uint,16,4) [] = { 0x8000, 0x8000, + 0x8000, 0x8000 }; +VECT_VAR_DECL(expected_max_sh_minus1,uint,32,2) [] = { 0x80000000, 0x80000000 }; +VECT_VAR_DECL(expected_max_sh_minus1,uint,64,1) [] = { 0x8000000000000000 }; +VECT_VAR_DECL(expected_max_sh_minus1,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_minus1,int,16,8) [] = { 0x4000, 0x4000, + 0x4000, 0x4000, + 0x4000, 0x4000, + 0x4000, 0x4000 }; +VECT_VAR_DECL(expected_max_sh_minus1,int,32,4) [] = { 0x40000000, 0x40000000, + 0x40000000, 0x40000000 }; +VECT_VAR_DECL(expected_max_sh_minus1,int,64,2) [] = { 0x4000000000000000, + 0x4000000000000000 }; +VECT_VAR_DECL(expected_max_sh_minus1,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_minus1,uint,16,8) [] = { 0x8000, 0x8000, + 0x8000, 0x8000, + 0x8000, 0x8000, + 0x8000, 0x8000 }; +VECT_VAR_DECL(expected_max_sh_minus1,uint,32,4) [] = { 0x80000000, 0x80000000, + 0x80000000, 0x80000000 }; +VECT_VAR_DECL(expected_max_sh_minus1,uint,64,2) [] = { 0x8000000000000000, + 0x8000000000000000 }; + +/* Expected results with max input value shifted by -3 to test + round_const. */ +VECT_VAR_DECL(expected_max_sh_minus3,int,8,8) [] = { 0x10, 0x10, 0x10, 0x10, + 0x10, 0x10, 0x10, 0x10 }; +VECT_VAR_DECL(expected_max_sh_minus3,int,16,4) [] = { 0x1000, 0x1000, + 0x1000, 0x1000 }; +VECT_VAR_DECL(expected_max_sh_minus3,int,32,2) [] = { 0x10000000, 0x10000000 }; +VECT_VAR_DECL(expected_max_sh_minus3,int,64,1) [] = { 0x1000000000000000 }; +VECT_VAR_DECL(expected_max_sh_minus3,uint,8,8) [] = { 0x20, 0x20, 0x20, 0x20, + 0x20, 0x20, 0x20, 0x20 }; +VECT_VAR_DECL(expected_max_sh_minus3,uint,16,4) [] = { 0x2000, 0x2000, + 0x2000, 0x2000 }; +VECT_VAR_DECL(expected_max_sh_minus3,uint,32,2) [] = { 0x20000000, 0x20000000 }; +VECT_VAR_DECL(expected_max_sh_minus3,uint,64,1) [] = { 0x2000000000000000 }; +VECT_VAR_DECL(expected_max_sh_minus3,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_minus3,int,16,8) [] = { 0x1000, 0x1000, + 0x1000, 0x1000, + 0x1000, 0x1000, + 0x1000, 0x1000 }; +VECT_VAR_DECL(expected_max_sh_minus3,int,32,4) [] = { 0x10000000, 0x10000000, + 0x10000000, 0x10000000 }; +VECT_VAR_DECL(expected_max_sh_minus3,int,64,2) [] = { 0x1000000000000000, + 0x1000000000000000 }; +VECT_VAR_DECL(expected_max_sh_minus3,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_minus3,uint,16,8) [] = { 0x2000, 0x2000, + 0x2000, 0x2000, + 0x2000, 0x2000, + 0x2000, 0x2000 }; +VECT_VAR_DECL(expected_max_sh_minus3,uint,32,4) [] = { 0x20000000, 0x20000000, + 0x20000000, 0x20000000 }; +VECT_VAR_DECL(expected_max_sh_minus3,uint,64,2) [] = { 0x2000000000000000, + 0x2000000000000000 }; + +/* Expected results with negative shift by vector width. */ +VECT_VAR_DECL(expected_max_sh_minus_width,int,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_minus_width,int,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_minus_width,int,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_minus_width,int,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_max_sh_minus_width,uint,8,8) [] = { 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_minus_width,uint,16,4) [] = { 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_minus_width,uint,32,2) [] = { 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_minus_width,uint,64,1) [] = { 0x1 }; +VECT_VAR_DECL(expected_max_sh_minus_width,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_minus_width,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_minus_width,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_minus_width,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_minus_width,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_minus_width,uint,16,8) [] = { 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_minus_width,uint,32,4) [] = { 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_minus_width,uint,64,2) [] = { 0x1, 0x1 }; + +/* Expected results with large shift amount. */ +VECT_VAR_DECL(expected_max_sh_large,int,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,int,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,int,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,int,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,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_large,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large,uint,64,2) [] = { 0x0, 0x0 }; + +/* Expected results with large negative shift amount. */ +VECT_VAR_DECL(expected_max_sh_large_neg,int,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,int,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,int,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,int,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,8,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,16,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,32,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,64,1) [] = { 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,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_large_neg,int,16,8) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,int,32,4) [] = { 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,int,64,2) [] = { 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,8,16) [] = { 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,16,8) [] = { 0x1, 0x1, 0x1, 0x1, + 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,32,4) [] = { 0x1, 0x1, 0x1, 0x1 }; +VECT_VAR_DECL(expected_max_sh_large_neg,uint,64,2) [] = { 0x1, 0x1 }; + +#define TEST_MSG "VRSHL/VRSHLQ" +void exec_vrshl (void) +{ + /* Basic test: v3=vrshl(v1,v2), then store the result. */ +#define TEST_VRSHL(T3, Q, T1, T2, W, N) \ + VECT_VAR(vector_res, T1, W, N) = \ + vrshl##Q##_##T2##W(VECT_VAR(vector, T1, W, N), \ + VECT_VAR(vector_shift, T3, W, N)); \ + 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); + + DECL_VARIABLE_SIGNED_VARIANTS(vector_shift); + + clean_results (); + + /* Fill input vector with 0, to check behavior on limits. */ + VDUP(vector, , int, s, 8, 8, 0); + VDUP(vector, , int, s, 16, 4, 0); + VDUP(vector, , int, s, 32, 2, 0); + VDUP(vector, , int, s, 64, 1, 0); + VDUP(vector, , uint, u, 8, 8, 0); + VDUP(vector, , uint, u, 16, 4, 0); + VDUP(vector, , uint, u, 32, 2, 0); + VDUP(vector, , uint, u, 64, 1, 0); + VDUP(vector, q, int, s, 8, 16, 0); + VDUP(vector, q, int, s, 16, 8, 0); + VDUP(vector, q, int, s, 32, 4, 0); + VDUP(vector, q, int, s, 64, 2, 0); + VDUP(vector, q, uint, u, 8, 16, 0); + VDUP(vector, q, uint, u, 16, 8, 0); + VDUP(vector, q, uint, u, 32, 4, 0); + VDUP(vector, q, uint, u, 64, 2, 0); + + /* Choose init value arbitrarily, will be used as shift amount. */ + /* Use values equal to one-less-than the type width to check + behaviour on limits. */ + VDUP(vector_shift, , int, s, 8, 8, 7); + VDUP(vector_shift, , int, s, 16, 4, 15); + VDUP(vector_shift, , int, s, 32, 2, 31); + VDUP(vector_shift, , int, s, 64, 1, 63); + VDUP(vector_shift, q, int, s, 8, 16, 7); + VDUP(vector_shift, q, int, s, 16, 8, 15); + VDUP(vector_shift, q, int, s, 32, 4, 31); + VDUP(vector_shift, q, int, s, 64, 2, 63); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#define CMT " (with input = 0)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_0, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_0, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_0, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_0, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_0, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_0, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_0, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_0, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_0, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_0, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_0, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_0, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_0, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_0, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_0, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_0, CMT); + + + /* Use negative shift amounts. */ + VDUP(vector_shift, , int, s, 8, 8, -1); + VDUP(vector_shift, , int, s, 16, 4, -2); + VDUP(vector_shift, , int, s, 32, 2, -3); + VDUP(vector_shift, , int, s, 64, 1, -4); + VDUP(vector_shift, q, int, s, 8, 16, -7); + VDUP(vector_shift, q, int, s, 16, 8, -11); + VDUP(vector_shift, q, int, s, 32, 4, -13); + VDUP(vector_shift, q, int, s, 64, 2, -20); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#define CMT " (input 0 and negative shift amount)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_0_sh_neg, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_0_sh_neg, CMT); + + + /* Test again, with predefined input values. */ + TEST_MACRO_ALL_VARIANTS_2_5(VLOAD, vector, buffer); + + /* Choose init value arbitrarily, will be used as shift amount. */ + VDUP(vector_shift, , int, s, 8, 8, 1); + VDUP(vector_shift, , int, s, 16, 4, 3); + VDUP(vector_shift, , int, s, 32, 2, 8); + VDUP(vector_shift, , int, s, 64, 1, -3); + VDUP(vector_shift, q, int, s, 8, 16, 10); + VDUP(vector_shift, q, int, s, 16, 8, 12); + VDUP(vector_shift, q, int, s, 32, 4, 32); + VDUP(vector_shift, q, int, s, 64, 2, 63); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#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 negative shift amounts. */ + VDUP(vector_shift, , int, s, 8, 8, -1); + VDUP(vector_shift, , int, s, 16, 4, -2); + VDUP(vector_shift, , int, s, 32, 2, -3); + VDUP(vector_shift, , int, s, 64, 1, -4); + VDUP(vector_shift, q, int, s, 8, 16, -7); + VDUP(vector_shift, q, int, s, 16, 8, -11); + VDUP(vector_shift, q, int, s, 32, 4, -13); + VDUP(vector_shift, q, int, s, 64, 2, -20); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#define CMT " (negative shift amount)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_sh_neg, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_sh_neg, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_sh_neg, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_sh_neg, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_sh_neg, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_sh_neg, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_sh_neg, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_sh_neg, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_sh_neg, CMT); + + /* Fill input vector with max value, to check behavior on limits. */ + 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 -1 shift amount to check overflow with round_const. */ + VDUP(vector_shift, , int, s, 8, 8, -1); + VDUP(vector_shift, , int, s, 16, 4, -1); + VDUP(vector_shift, , int, s, 32, 2, -1); + VDUP(vector_shift, , int, s, 64, 1, -1); + VDUP(vector_shift, q, int, s, 8, 16, -1); + VDUP(vector_shift, q, int, s, 16, 8, -1); + VDUP(vector_shift, q, int, s, 32, 4, -1); + VDUP(vector_shift, q, int, s, 64, 2, -1); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#define CMT " (max input, shift by -1)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_minus1, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_minus1, CMT); + + /* Use -3 shift amount to check overflow with round_const. */ + VDUP(vector_shift, , int, s, 8, 8, -3); + VDUP(vector_shift, , int, s, 16, 4, -3); + VDUP(vector_shift, , int, s, 32, 2, -3); + VDUP(vector_shift, , int, s, 64, 1, -3); + VDUP(vector_shift, q, int, s, 8, 16, -3); + VDUP(vector_shift, q, int, s, 16, 8, -3); + VDUP(vector_shift, q, int, s, 32, 4, -3); + VDUP(vector_shift, q, int, s, 64, 2, -3); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#define CMT " (check rounding constant: max input, shift by -3)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_minus3, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_minus3, CMT); + + + /* Use negative shift amount as large as input vector width. */ + VDUP(vector_shift, , int, s, 8, 8, -8); + VDUP(vector_shift, , int, s, 16, 4, -16); + VDUP(vector_shift, , int, s, 32, 2, -32); + VDUP(vector_shift, , int, s, 64, 1, -64); + VDUP(vector_shift, q, int, s, 8, 16, -8); + VDUP(vector_shift, q, int, s, 16, 8, -16); + VDUP(vector_shift, q, int, s, 32, 4, -32); + VDUP(vector_shift, q, int, s, 64, 2, -64); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#define CMT " (max input, right shift by vector width)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_minus_width, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_minus_width, CMT); + + + /* Test large shift amount. */ + VDUP(vector_shift, , int, s, 8, 8, 10); + VDUP(vector_shift, , int, s, 16, 4, 20); + VDUP(vector_shift, , int, s, 32, 2, 33); + VDUP(vector_shift, , int, s, 64, 1, 65); + VDUP(vector_shift, q, int, s, 8, 16, 9); + VDUP(vector_shift, q, int, s, 16, 8, 16); + VDUP(vector_shift, q, int, s, 32, 4, 32); + VDUP(vector_shift, q, int, s, 64, 2, 64); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#define CMT " (max input, large shift amount)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_large, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_large, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_large, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_large, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_large, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_large, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_large, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_large, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_large, CMT); + + + /* Test large negative shift amount. */ + VDUP(vector_shift, , int, s, 8, 8, -10); + VDUP(vector_shift, , int, s, 16, 4, -20); + VDUP(vector_shift, , int, s, 32, 2, -33); + VDUP(vector_shift, , int, s, 64, 1, -65); + VDUP(vector_shift, q, int, s, 8, 16, -9); + VDUP(vector_shift, q, int, s, 16, 8, -16); + VDUP(vector_shift, q, int, s, 32, 4, -32); + VDUP(vector_shift, q, int, s, 64, 2, -64); + + TEST_MACRO_ALL_VARIANTS_1_5(TEST_VRSHL, int); + +#undef CMT +#define CMT " (max input, large negative shift amount)" + CHECK(TEST_MSG, int, 8, 8, PRIx8, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, int, 16, 4, PRIx16, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, int, 32, 2, PRIx32, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, int, 64, 1, PRIx64, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 8, 8, PRIx8, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 16, 4, PRIx16, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 32, 2, PRIx32, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 64, 1, PRIx64, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, int, 8, 16, PRIx8, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, int, 16, 8, PRIx16, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, int, 32, 4, PRIx32, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, int, 64, 2, PRIx64, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 8, 16, PRIx8, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 16, 8, PRIx16, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 32, 4, PRIx32, expected_max_sh_large_neg, CMT); + CHECK(TEST_MSG, uint, 64, 2, PRIx64, expected_max_sh_large_neg, CMT); +} + +int main (void) +{ + exec_vrshl (); + return 0; +}