From patchwork Fri Nov 3 12:12:20 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Joe Ramsay X-Patchwork-Id: 1858881 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=armh.onmicrosoft.com header.i=@armh.onmicrosoft.com header.a=rsa-sha256 header.s=selector2-armh-onmicrosoft-com header.b=qSzli14W; dkim=pass (1024-bit key) header.d=armh.onmicrosoft.com header.i=@armh.onmicrosoft.com header.a=rsa-sha256 header.s=selector2-armh-onmicrosoft-com header.b=qSzli14W; dkim-atps=neutral Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=sourceware.org (client-ip=8.43.85.97; helo=server2.sourceware.org; envelope-from=libc-alpha-bounces+incoming=patchwork.ozlabs.org@sourceware.org; receiver=patchwork.ozlabs.org) Received: from server2.sourceware.org (server2.sourceware.org [8.43.85.97]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (secp384r1) server-digest SHA384) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4SMKQR3phBz1yQ5 for ; Fri, 3 Nov 2023 23:13:16 +1100 (AEDT) Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 351DC3858418 for ; Fri, 3 Nov 2023 12:13:14 +0000 (GMT) X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from EUR01-VE1-obe.outbound.protection.outlook.com (mail-ve1eur01on2078.outbound.protection.outlook.com [40.107.14.78]) by sourceware.org (Postfix) with ESMTPS id 43FBA3858C41 for ; Fri, 3 Nov 2023 12:12:53 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 43FBA3858C41 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=arm.com Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=arm.com ARC-Filter: OpenARC Filter v1.0.0 sourceware.org 43FBA3858C41 Authentication-Results: server2.sourceware.org; arc=pass smtp.remote-ip=40.107.14.78 ARC-Seal: i=3; a=rsa-sha256; d=sourceware.org; s=key; t=1699013583; cv=pass; b=p6IqDGfTnoeAjWOFjN5Paju1CK0FpVZ2qCIjw7yoIBYrJVq7AMsxZojZo3R7PlBntnjcXpg/7iMg0kn1csLLOBw+lSbEyuVZmsUbiMSYREObMYeesjX4YJuONQwOY2esKp7iwNwr5mX2mlZQBGQm05I4va3pdkrzr28482Nb/ZE= ARC-Message-Signature: i=3; a=rsa-sha256; d=sourceware.org; s=key; t=1699013583; c=relaxed/simple; bh=txQkj4mlcAv/ed189XLBb5X8ZM+5korTVL/1aQ26za4=; h=DKIM-Signature:DKIM-Signature:From:To:Subject:Date:Message-ID: MIME-Version; b=rd4GIGCRPSedwk6PfdJRWwRLPxY8e9xCsPcTbEHW3nSLDHIeDHtaMmC+yqLSOStpvHc5dWeiChWAGxFI21w8jCrSWfgRhoK/mGZejzhb3IjwIcYc4QiNjCmdIcQ7Udgo83U5K5wiYnMrIHHHXWO8dM1uwIsnvMnUy+iWAl51fAI= ARC-Authentication-Results: i=3; server2.sourceware.org ARC-Seal: i=2; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=pass; b=R9bNEY0khjhYuaYGf07mBmw24mtTXnM6NHRt+h3C5INuO+mAJzv/WuvVFPDKVLx92YMZN9gu/rdWvCPruKofhvfRCegE7O6nyc5PPJ3yL1n9XD3fmvVKz11otf/1byVUDKICUOadgV+wDCgb2mISIoHBjngj1nCEL97WaSZTB6FpjZsX1WuRqAOImeC+O7QKrA6VGBfU4tnXcH2W67rp2w/wyiUh8Pb1YFNhhcaEg7S7mvvk30Z7YzutFkR7vY2O8wtLudUken8AXdd4KYgQZpdoTaW7kDf77r3Sx8Fo+tahDwEkgfZ2DRgboVhHcgFpwFJ9FOGjmZoiOMt2KrchKw== ARC-Message-Signature: i=2; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=eUVXUq8CDOSNcO6S0Ej+vnmdg8vDsA6FHUTvY+COHNc=; b=HgiiVRDKmxx+uH27tjZuEAAtGwA02vPyaLbrpK+uDZMcb3zkqsZmPgGObOQtBCCAy8fddXvt2BhzltbdBhSNre/KZ+hqFLvx/arJZs5c2y9JILZxDTSoAvyRL6NQm43faK9y3QlY/K7z53+iTCqfPTxdhAIeF9Fy+WF4iGBR0l6nfScL98yb1IcJ/O+7lNBERc32uXlCbwYfhZzOwGXPMu4YSYvPqKeFdX9bPC+Pq7GXvH+I3u0lbs4Hgzn7gYEbg6QkMphZFyepUhperrdQRh3SyJkpLPa6Tj6/OuXN/3ffeniD/3X7v7g7re2K1B50npWilDFGFnxSEESAJ8N+Sw== ARC-Authentication-Results: i=2; mx.microsoft.com 1; spf=pass (sender ip is 63.35.35.123) smtp.rcpttodomain=sourceware.org smtp.mailfrom=arm.com; dmarc=pass (p=none sp=none pct=100) action=none header.from=arm.com; dkim=pass (signature was verified) header.d=armh.onmicrosoft.com; arc=pass (0 oda=1 ltdi=1 spf=[1, 1, smtp.mailfrom=arm.com] dmarc=[1, 1, header.from=arm.com]) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=armh.onmicrosoft.com; s=selector2-armh-onmicrosoft-com; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=eUVXUq8CDOSNcO6S0Ej+vnmdg8vDsA6FHUTvY+COHNc=; b=qSzli14WziNHjGHhOolNp/cnH4k+QGXRg6iLr7tGzMXkMr3ezmGiMaMrCcq5YB+A/OkvDnA1DBT/rONyI3I/MRyGl561xThZDRn12YelIxnWRwZcANkPML1Dk2p6qvab436lMMB6PJQK6ggPHwL6R/TdvDK1qAcuvqRqrWlCQeo= Received: from AS9PR06CA0085.eurprd06.prod.outlook.com (2603:10a6:20b:464::15) by AM9PR08MB6658.eurprd08.prod.outlook.com (2603:10a6:20b:303::14) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6954.21; Fri, 3 Nov 2023 12:12:48 +0000 Received: from AMS0EPF000001B6.eurprd05.prod.outlook.com (2603:10a6:20b:464:cafe::a) by AS9PR06CA0085.outlook.office365.com (2603:10a6:20b:464::15) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6954.21 via Frontend Transport; Fri, 3 Nov 2023 12:12:48 +0000 X-MS-Exchange-Authentication-Results: spf=pass (sender IP is 63.35.35.123) smtp.mailfrom=arm.com; dkim=pass (signature was verified) header.d=armh.onmicrosoft.com;dmarc=pass action=none header.from=arm.com; Received-SPF: Pass (protection.outlook.com: domain of arm.com designates 63.35.35.123 as permitted sender) receiver=protection.outlook.com; client-ip=63.35.35.123; helo=64aa7808-outbound-1.mta.getcheckrecipient.com; pr=C Received: from 64aa7808-outbound-1.mta.getcheckrecipient.com (63.35.35.123) by AMS0EPF000001B6.mail.protection.outlook.com (10.167.16.170) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6954.19 via Frontend Transport; Fri, 3 Nov 2023 12:12:47 +0000 Received: ("Tessian outbound e243565b0037:v228"); Fri, 03 Nov 2023 12:12:47 +0000 X-CheckRecipientChecked: true X-CR-MTA-CID: 089f874b6966afc7 X-CR-MTA-TID: 64aa7808 Received: from d47a92e24c38.1 by 64aa7808-outbound-1.mta.getcheckrecipient.com id FFCCCF96-6A4E-4DF5-87F8-0BA8680485AF.1; Fri, 03 Nov 2023 12:12:40 +0000 Received: from EUR04-HE1-obe.outbound.protection.outlook.com by 64aa7808-outbound-1.mta.getcheckrecipient.com with ESMTPS id d47a92e24c38.1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384); Fri, 03 Nov 2023 12:12:40 +0000 ARC-Seal: i=1; a=rsa-sha256; s=arcselector9901; d=microsoft.com; cv=none; b=ocke6Ox0wggs/TexqoNV9u79NGHvwN4mTQmK0UZ+KS5imo/9iTpDHW7nXle+6eoaasw48/cjPysGMoPYXggtBKYe9vzu6dEOJ6JfXHzmuajoFaxRB8WqVfYXvA4ihtDotrf0aZpwV3k1bKfnr8pZK+mbkKghPxIe1omMMI2zCILo8R5JHzBcaXqF6BrAOrBNEEObq2NxxWuv7sX5MF1EMzjvHGu8nq+Z0nEQ7teRtQGTXiJgPopjtJg3SjI8YwyjJ/29+/Icj0FjM9CybPGv3HsOhNGIwzuuhwqt4/aiiAqg/vjDUjUvbhADzOKBu+IFUkNhXZQmjFlpT6ujwxvXWA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=microsoft.com; s=arcselector9901; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-AntiSpam-MessageData-ChunkCount:X-MS-Exchange-AntiSpam-MessageData-0:X-MS-Exchange-AntiSpam-MessageData-1; bh=eUVXUq8CDOSNcO6S0Ej+vnmdg8vDsA6FHUTvY+COHNc=; b=ApzNRHzQtwoyDvWS52vI/27FRRRS6tryIQOGn2n10Ck3eN3GhBNo+5PV6eEbJd7vx2eDn/lY62qVuqGY8P1+qMgFlU9ONKyJM8leNaltlCoXo+gCK5+gqsY4qkfVAnTPzdnmbsN36/UPgW9gswH65gRUE+bW4AuU+zJGtiHbDs5PkWFvoRu9IV/Q9IqsOLnmch6bLV1WGWeJwufZvNzLEc/WDi7QxxkwOHUHmG0e7b/i95sb4UEoxHIav1R1UCl44o2UtwAMqHNKheO/IbbUXvt2DdIWhDBqDbtLYCZr1UJ2wdjJqusik+4Z9/Z4FYMw3Q6hW1Da7FPrY40erCFBDw== ARC-Authentication-Results: i=1; mx.microsoft.com 1; spf=pass (sender ip is 40.67.248.234) smtp.rcpttodomain=sourceware.org smtp.mailfrom=arm.com; dmarc=pass (p=none sp=none pct=100) action=none header.from=arm.com; dkim=none (message not signed); arc=none (0) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=armh.onmicrosoft.com; s=selector2-armh-onmicrosoft-com; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version:X-MS-Exchange-SenderADCheck; bh=eUVXUq8CDOSNcO6S0Ej+vnmdg8vDsA6FHUTvY+COHNc=; b=qSzli14WziNHjGHhOolNp/cnH4k+QGXRg6iLr7tGzMXkMr3ezmGiMaMrCcq5YB+A/OkvDnA1DBT/rONyI3I/MRyGl561xThZDRn12YelIxnWRwZcANkPML1Dk2p6qvab436lMMB6PJQK6ggPHwL6R/TdvDK1qAcuvqRqrWlCQeo= Received: from AS9PR04CA0071.eurprd04.prod.outlook.com (2603:10a6:20b:48b::20) by AM0PR08MB5329.eurprd08.prod.outlook.com (2603:10a6:208:185::19) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6954.21; Fri, 3 Nov 2023 12:12:38 +0000 Received: from AMS0EPF000001B5.eurprd05.prod.outlook.com (2603:10a6:20b:48b:cafe::c8) by AS9PR04CA0071.outlook.office365.com (2603:10a6:20b:48b::20) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.20.6954.22 via Frontend Transport; Fri, 3 Nov 2023 12:12:38 +0000 X-MS-Exchange-Authentication-Results: spf=pass (sender IP is 40.67.248.234) smtp.mailfrom=arm.com; dkim=none (message not signed) header.d=none;dmarc=pass action=none header.from=arm.com; Received-SPF: Pass (protection.outlook.com: domain of arm.com designates 40.67.248.234 as permitted sender) receiver=protection.outlook.com; client-ip=40.67.248.234; helo=nebula.arm.com; pr=C Received: from nebula.arm.com (40.67.248.234) by AMS0EPF000001B5.mail.protection.outlook.com (10.167.16.169) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.20.6954.19 via Frontend Transport; Fri, 3 Nov 2023 12:12:38 +0000 Received: from AZ-NEU-EX04.Arm.com (10.251.24.32) by AZ-NEU-EX03.Arm.com (10.251.24.31) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2507.32; Fri, 3 Nov 2023 12:12:36 +0000 Received: from vcn-man-apps.manchester.arm.com (10.32.108.22) by mail.arm.com (10.251.24.32) with Microsoft SMTP Server id 15.1.2507.32 via Frontend Transport; Fri, 3 Nov 2023 12:12:35 +0000 From: Joe Ramsay To: CC: Joe Ramsay Subject: [PATCH 2/6] aarch64: Add vector implementations of acos routines Date: Fri, 3 Nov 2023 12:12:20 +0000 Message-ID: <20231103121224.16835-2-Joe.Ramsay@arm.com> X-Mailer: git-send-email 2.27.0 In-Reply-To: <20231103121224.16835-1-Joe.Ramsay@arm.com> References: <20231103121224.16835-1-Joe.Ramsay@arm.com> MIME-Version: 1.0 X-EOPAttributedMessage: 1 X-MS-TrafficTypeDiagnostic: AMS0EPF000001B5:EE_|AM0PR08MB5329:EE_|AMS0EPF000001B6:EE_|AM9PR08MB6658:EE_ X-MS-Office365-Filtering-Correlation-Id: 581df5b7-dc9e-4599-7858-08dbdc663197 x-checkrecipientrouted: true NoDisclaimer: true X-MS-Exchange-SenderADCheck: 1 X-MS-Exchange-AntiSpam-Relay: 0 X-Microsoft-Antispam-Untrusted: BCL:0; X-Microsoft-Antispam-Message-Info-Original: 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 X-Forefront-Antispam-Report-Untrusted: CIP:40.67.248.234; CTRY:IE; LANG:en; SCL:1; SRV:; IPV:NLI; SFV:NSPM; H:nebula.arm.com; PTR:InfoDomainNonexistent; CAT:NONE; SFS:(13230031)(4636009)(346002)(39860400002)(376002)(396003)(136003)(230922051799003)(186009)(82310400011)(1800799009)(451199024)(64100799003)(40470700004)(36840700001)(46966006)(5660300002)(1076003)(2616005)(478600001)(26005)(426003)(6666004)(336012)(7696005)(86362001)(83380400001)(70586007)(356005)(70206006)(316002)(30864003)(6916009)(41300700001)(82740400003)(81166007)(40460700003)(2906002)(40480700001)(4326008)(8936002)(8676002)(36756003)(47076005)(36860700001)(2004002)(36900700001); DIR:OUT; SFP:1101; X-MS-Exchange-Transport-CrossTenantHeadersStamped: AM0PR08MB5329 X-MS-Exchange-Transport-CrossTenantHeadersStripped: AMS0EPF000001B6.eurprd05.prod.outlook.com X-MS-PublicTrafficType: Email X-MS-Office365-Filtering-Correlation-Id-Prvs: 0980f24d-c3d5-4e86-6fe5-08dbdc662bbc X-Microsoft-Antispam: BCL:0; X-Microsoft-Antispam-Message-Info: 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 X-Forefront-Antispam-Report: CIP:63.35.35.123; CTRY:IE; LANG:en; SCL:1; SRV:; IPV:CAL; SFV:NSPM; H:64aa7808-outbound-1.mta.getcheckrecipient.com; PTR:ec2-63-35-35-123.eu-west-1.compute.amazonaws.com; CAT:NONE; SFS:(13230031)(4636009)(346002)(396003)(136003)(39860400002)(376002)(230922051799003)(451199024)(82310400011)(1800799009)(186009)(64100799003)(36840700001)(46966006)(40470700004)(82740400003)(83380400001)(36860700001)(47076005)(40480700001)(40460700003)(41300700001)(70206006)(70586007)(478600001)(316002)(6916009)(4326008)(8676002)(2906002)(6666004)(30864003)(8936002)(5660300002)(7696005)(86362001)(1076003)(36756003)(26005)(336012)(81166007)(426003)(2616005)(2004002); DIR:OUT; SFP:1101; X-OriginatorOrg: arm.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 03 Nov 2023 12:12:47.9217 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: 581df5b7-dc9e-4599-7858-08dbdc663197 X-MS-Exchange-CrossTenant-Id: f34e5979-57d9-4aaa-ad4d-b122a662184d X-MS-Exchange-CrossTenant-OriginalAttributedTenantConnectingIp: TenantId=f34e5979-57d9-4aaa-ad4d-b122a662184d; Ip=[63.35.35.123]; Helo=[64aa7808-outbound-1.mta.getcheckrecipient.com] X-MS-Exchange-CrossTenant-AuthSource: AMS0EPF000001B6.eurprd05.prod.outlook.com X-MS-Exchange-CrossTenant-AuthAs: Anonymous X-MS-Exchange-CrossTenant-FromEntityHeader: HybridOnPrem X-MS-Exchange-Transport-CrossTenantHeadersStamped: AM9PR08MB6658 X-Spam-Status: No, score=-12.4 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, FORGED_SPF_HELO, GIT_PATCH_0, KAM_DMARC_NONE, KAM_SHORT, RCVD_IN_DNSWL_NONE, RCVD_IN_MSPIKE_H2, SPF_HELO_PASS, SPF_NONE, TXREP, T_SCC_BODY_TEXT_LINE, UNPARSEABLE_RELAY autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.30 Precedence: list List-Id: Libc-alpha mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: libc-alpha-bounces+incoming=patchwork.ozlabs.org@sourceware.org --- Thanks, Joe sysdeps/aarch64/fpu/Makefile | 3 +- sysdeps/aarch64/fpu/Versions | 4 + sysdeps/aarch64/fpu/acos_advsimd.c | 122 ++++++++++++++++++ sysdeps/aarch64/fpu/acos_sve.c | 93 +++++++++++++ sysdeps/aarch64/fpu/acosf_advsimd.c | 113 ++++++++++++++++ sysdeps/aarch64/fpu/acosf_sve.c | 86 ++++++++++++ sysdeps/aarch64/fpu/bits/math-vector.h | 4 + .../fpu/test-double-advsimd-wrappers.c | 1 + .../aarch64/fpu/test-double-sve-wrappers.c | 1 + .../aarch64/fpu/test-float-advsimd-wrappers.c | 1 + sysdeps/aarch64/fpu/test-float-sve-wrappers.c | 1 + sysdeps/aarch64/libm-test-ulps | 8 ++ .../unix/sysv/linux/aarch64/libmvec.abilist | 4 + 13 files changed, 440 insertions(+), 1 deletion(-) create mode 100644 sysdeps/aarch64/fpu/acos_advsimd.c create mode 100644 sysdeps/aarch64/fpu/acos_sve.c create mode 100644 sysdeps/aarch64/fpu/acosf_advsimd.c create mode 100644 sysdeps/aarch64/fpu/acosf_sve.c Reviewed-by: Szabolcs Nagy diff --git a/sysdeps/aarch64/fpu/Makefile b/sysdeps/aarch64/fpu/Makefile index d7c0bd2ed5..606fdd804f 100644 --- a/sysdeps/aarch64/fpu/Makefile +++ b/sysdeps/aarch64/fpu/Makefile @@ -1,4 +1,5 @@ -libmvec-supported-funcs = asin \ +libmvec-supported-funcs = acos \ + asin \ cos \ exp \ exp10 \ diff --git a/sysdeps/aarch64/fpu/Versions b/sysdeps/aarch64/fpu/Versions index 0f365a1e2e..1037cd92bd 100644 --- a/sysdeps/aarch64/fpu/Versions +++ b/sysdeps/aarch64/fpu/Versions @@ -18,6 +18,10 @@ libmvec { _ZGVsMxv_sinf; } GLIBC_2.39 { + _ZGVnN4v_acosf; + _ZGVnN2v_acos; + _ZGVsMxv_acosf; + _ZGVsMxv_acos; _ZGVnN4v_asinf; _ZGVnN2v_asin; _ZGVsMxv_asinf; diff --git a/sysdeps/aarch64/fpu/acos_advsimd.c b/sysdeps/aarch64/fpu/acos_advsimd.c new file mode 100644 index 0000000000..3121cf66b1 --- /dev/null +++ b/sysdeps/aarch64/fpu/acos_advsimd.c @@ -0,0 +1,122 @@ +/* Double-precision AdvSIMD inverse cos + + Copyright (C) 2023 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + . */ + +#include "v_math.h" +#include "poly_advsimd_f64.h" + +static const struct data +{ + float64x2_t poly[12]; + float64x2_t pi, pi_over_2; + uint64x2_t abs_mask; +} data = { + /* Polynomial approximation of (asin(sqrt(x)) - sqrt(x)) / (x * sqrt(x)) + on [ 0x1p-106, 0x1p-2 ], relative error: 0x1.c3d8e169p-57. */ + .poly = { V2 (0x1.555555555554ep-3), V2 (0x1.3333333337233p-4), + V2 (0x1.6db6db67f6d9fp-5), V2 (0x1.f1c71fbd29fbbp-6), + V2 (0x1.6e8b264d467d6p-6), V2 (0x1.1c5997c357e9dp-6), + V2 (0x1.c86a22cd9389dp-7), V2 (0x1.856073c22ebbep-7), + V2 (0x1.fd1151acb6bedp-8), V2 (0x1.087182f799c1dp-6), + V2 (-0x1.6602748120927p-7), V2 (0x1.cfa0dd1f9478p-6), }, + .pi = V2 (0x1.921fb54442d18p+1), + .pi_over_2 = V2 (0x1.921fb54442d18p+0), + .abs_mask = V2 (0x7fffffffffffffff), +}; + +#define AllMask v_u64 (0xffffffffffffffff) +#define Oneu (0x3ff0000000000000) +#define Small (0x3e50000000000000) /* 2^-53. */ + +#if WANT_SIMD_EXCEPT +static float64x2_t VPCS_ATTR NOINLINE +special_case (float64x2_t x, float64x2_t y, uint64x2_t special) +{ + return v_call_f64 (acos, x, y, special); +} +#endif + +/* Double-precision implementation of vector acos(x). + + For |x| < Small, approximate acos(x) by pi/2 - x. Small = 2^-53 for correct + rounding. + If WANT_SIMD_EXCEPT = 0, Small = 0 and we proceed with the following + approximation. + + For |x| in [Small, 0.5], use an order 11 polynomial P such that the final + approximation of asin is an odd polynomial: + + acos(x) ~ pi/2 - (x + x^3 P(x^2)). + + The largest observed error in this region is 1.18 ulps, + _ZGVnN2v_acos (0x1.fbab0a7c460f6p-2) got 0x1.0d54d1985c068p+0 + want 0x1.0d54d1985c069p+0. + + For |x| in [0.5, 1.0], use same approximation with a change of variable + + acos(x) = y + y * z * P(z), with z = (1-x)/2 and y = sqrt(z). + + The largest observed error in this region is 1.52 ulps, + _ZGVnN2v_acos (0x1.23d362722f591p-1) got 0x1.edbbedf8a7d6ep-1 + want 0x1.edbbedf8a7d6cp-1. */ +float64x2_t VPCS_ATTR V_NAME_D1 (acos) (float64x2_t x) +{ + const struct data *d = ptr_barrier (&data); + + float64x2_t ax = vabsq_f64 (x); + +#if WANT_SIMD_EXCEPT + /* A single comparison for One, Small and QNaN. */ + uint64x2_t special + = vcgtq_u64 (vsubq_u64 (vreinterpretq_u64_f64 (ax), v_u64 (Small)), + v_u64 (Oneu - Small)); + if (__glibc_unlikely (v_any_u64 (special))) + return special_case (x, x, AllMask); +#endif + + uint64x2_t a_le_half = vcleq_f64 (ax, v_f64 (0.5)); + + /* Evaluate polynomial Q(x) = z + z * z2 * P(z2) with + z2 = x ^ 2 and z = |x| , if |x| < 0.5 + z2 = (1 - |x|) / 2 and z = sqrt(z2), if |x| >= 0.5. */ + float64x2_t z2 = vbslq_f64 (a_le_half, vmulq_f64 (x, x), + vfmaq_f64 (v_f64 (0.5), v_f64 (-0.5), ax)); + float64x2_t z = vbslq_f64 (a_le_half, ax, vsqrtq_f64 (z2)); + + /* Use a single polynomial approximation P for both intervals. */ + float64x2_t z4 = vmulq_f64 (z2, z2); + float64x2_t z8 = vmulq_f64 (z4, z4); + float64x2_t z16 = vmulq_f64 (z8, z8); + float64x2_t p = v_estrin_11_f64 (z2, z4, z8, z16, d->poly); + + /* Finalize polynomial: z + z * z2 * P(z2). */ + p = vfmaq_f64 (z, vmulq_f64 (z, z2), p); + + /* acos(|x|) = pi/2 - sign(x) * Q(|x|), for |x| < 0.5 + = 2 Q(|x|) , for 0.5 < x < 1.0 + = pi - 2 Q(|x|) , for -1.0 < x < -0.5. */ + float64x2_t y = vbslq_f64 (d->abs_mask, p, x); + + uint64x2_t is_neg = vcltzq_f64 (x); + float64x2_t off = vreinterpretq_f64_u64 ( + vandq_u64 (is_neg, vreinterpretq_u64_f64 (d->pi))); + float64x2_t mul = vbslq_f64 (a_le_half, v_f64 (-1.0), v_f64 (2.0)); + float64x2_t add = vbslq_f64 (a_le_half, d->pi_over_2, off); + + return vfmaq_f64 (add, mul, y); +} diff --git a/sysdeps/aarch64/fpu/acos_sve.c b/sysdeps/aarch64/fpu/acos_sve.c new file mode 100644 index 0000000000..1138a04e73 --- /dev/null +++ b/sysdeps/aarch64/fpu/acos_sve.c @@ -0,0 +1,93 @@ +/* Double-precision SVE inverse cos + + Copyright (C) 2023 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + . */ + +#include "sv_math.h" +#include "poly_sve_f64.h" + +static const struct data +{ + float64_t poly[12]; + float64_t pi, pi_over_2; +} data = { + /* Polynomial approximation of (asin(sqrt(x)) - sqrt(x)) / (x * sqrt(x)) + on [ 0x1p-106, 0x1p-2 ], relative error: 0x1.c3d8e169p-57. */ + .poly = { 0x1.555555555554ep-3, 0x1.3333333337233p-4, 0x1.6db6db67f6d9fp-5, + 0x1.f1c71fbd29fbbp-6, 0x1.6e8b264d467d6p-6, 0x1.1c5997c357e9dp-6, + 0x1.c86a22cd9389dp-7, 0x1.856073c22ebbep-7, 0x1.fd1151acb6bedp-8, + 0x1.087182f799c1dp-6, -0x1.6602748120927p-7, 0x1.cfa0dd1f9478p-6, }, + .pi = 0x1.921fb54442d18p+1, + .pi_over_2 = 0x1.921fb54442d18p+0, +}; + +/* Double-precision SVE implementation of vector acos(x). + + For |x| in [0, 0.5], use an order 11 polynomial P such that the final + approximation of asin is an odd polynomial: + + acos(x) ~ pi/2 - (x + x^3 P(x^2)). + + The largest observed error in this region is 1.18 ulps, + _ZGVsMxv_acos (0x1.fbc5fe28ee9e3p-2) got 0x1.0d4d0f55667f6p+0 + want 0x1.0d4d0f55667f7p+0. + + For |x| in [0.5, 1.0], use same approximation with a change of variable + + acos(x) = y + y * z * P(z), with z = (1-x)/2 and y = sqrt(z). + + The largest observed error in this region is 1.52 ulps, + _ZGVsMxv_acos (0x1.24024271a500ap-1) got 0x1.ed82df4243f0dp-1 + want 0x1.ed82df4243f0bp-1. */ +svfloat64_t SV_NAME_D1 (acos) (svfloat64_t x, const svbool_t pg) +{ + const struct data *d = ptr_barrier (&data); + + svuint64_t sign = svand_x (pg, svreinterpret_u64 (x), 0x8000000000000000); + svfloat64_t ax = svabs_x (pg, x); + + svbool_t a_gt_half = svacgt (pg, x, 0.5); + + /* Evaluate polynomial Q(x) = z + z * z2 * P(z2) with + z2 = x ^ 2 and z = |x| , if |x| < 0.5 + z2 = (1 - |x|) / 2 and z = sqrt(z2), if |x| >= 0.5. */ + svfloat64_t z2 = svsel (a_gt_half, svmls_x (pg, sv_f64 (0.5), ax, 0.5), + svmul_x (pg, x, x)); + svfloat64_t z = svsqrt_m (ax, a_gt_half, z2); + + /* Use a single polynomial approximation P for both intervals. */ + svfloat64_t z4 = svmul_x (pg, z2, z2); + svfloat64_t z8 = svmul_x (pg, z4, z4); + svfloat64_t z16 = svmul_x (pg, z8, z8); + svfloat64_t p = sv_estrin_11_f64_x (pg, z2, z4, z8, z16, d->poly); + + /* Finalize polynomial: z + z * z2 * P(z2). */ + p = svmla_x (pg, z, svmul_x (pg, z, z2), p); + + /* acos(|x|) = pi/2 - sign(x) * Q(|x|), for |x| < 0.5 + = 2 Q(|x|) , for 0.5 < x < 1.0 + = pi - 2 Q(|x|) , for -1.0 < x < -0.5. */ + svfloat64_t y + = svreinterpret_f64 (svorr_x (pg, svreinterpret_u64 (p), sign)); + + svbool_t is_neg = svcmplt (pg, x, 0.0); + svfloat64_t off = svdup_f64_z (is_neg, d->pi); + svfloat64_t mul = svsel (a_gt_half, sv_f64 (2.0), sv_f64 (-1.0)); + svfloat64_t add = svsel (a_gt_half, off, sv_f64 (d->pi_over_2)); + + return svmla_x (pg, add, mul, y); +} diff --git a/sysdeps/aarch64/fpu/acosf_advsimd.c b/sysdeps/aarch64/fpu/acosf_advsimd.c new file mode 100644 index 0000000000..7d39e9b805 --- /dev/null +++ b/sysdeps/aarch64/fpu/acosf_advsimd.c @@ -0,0 +1,113 @@ +/* Single-precision AdvSIMD inverse cos + + Copyright (C) 2023 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + . */ + +#include "v_math.h" +#include "poly_advsimd_f32.h" + +static const struct data +{ + float32x4_t poly[5]; + float32x4_t pi_over_2f, pif; +} data = { + /* Polynomial approximation of (asin(sqrt(x)) - sqrt(x)) / (x * sqrt(x)) on + [ 0x1p-24 0x1p-2 ] order = 4 rel error: 0x1.00a23bbp-29 . */ + .poly = { V4 (0x1.55555ep-3), V4 (0x1.33261ap-4), V4 (0x1.70d7dcp-5), + V4 (0x1.b059dp-6), V4 (0x1.3af7d8p-5) }, + .pi_over_2f = V4 (0x1.921fb6p+0f), + .pif = V4 (0x1.921fb6p+1f), +}; + +#define AbsMask 0x7fffffff +#define Half 0x3f000000 +#define One 0x3f800000 +#define Small 0x32800000 /* 2^-26. */ + +#if WANT_SIMD_EXCEPT +static float32x4_t VPCS_ATTR NOINLINE +special_case (float32x4_t x, float32x4_t y, uint32x4_t special) +{ + return v_call_f32 (acosf, x, y, special); +} +#endif + +/* Single-precision implementation of vector acos(x). + + For |x| < Small, approximate acos(x) by pi/2 - x. Small = 2^-26 for correct + rounding. + If WANT_SIMD_EXCEPT = 0, Small = 0 and we proceed with the following + approximation. + + For |x| in [Small, 0.5], use order 4 polynomial P such that the final + approximation of asin is an odd polynomial: + + acos(x) ~ pi/2 - (x + x^3 P(x^2)). + + The largest observed error in this region is 1.26 ulps, + _ZGVnN4v_acosf (0x1.843bfcp-2) got 0x1.2e934cp+0 want 0x1.2e934ap+0. + + For |x| in [0.5, 1.0], use same approximation with a change of variable + + acos(x) = y + y * z * P(z), with z = (1-x)/2 and y = sqrt(z). + + The largest observed error in this region is 1.32 ulps, + _ZGVnN4v_acosf (0x1.15ba56p-1) got 0x1.feb33p-1 + want 0x1.feb32ep-1. */ +float32x4_t VPCS_ATTR V_NAME_F1 (acos) (float32x4_t x) +{ + const struct data *d = ptr_barrier (&data); + + uint32x4_t ix = vreinterpretq_u32_f32 (x); + uint32x4_t ia = vandq_u32 (ix, v_u32 (AbsMask)); + +#if WANT_SIMD_EXCEPT + /* A single comparison for One, Small and QNaN. */ + uint32x4_t special + = vcgtq_u32 (vsubq_u32 (ia, v_u32 (Small)), v_u32 (One - Small)); + if (__glibc_unlikely (v_any_u32 (special))) + return special_case (x, x, v_u32 (0xffffffff)); +#endif + + float32x4_t ax = vreinterpretq_f32_u32 (ia); + uint32x4_t a_le_half = vcleq_u32 (ia, v_u32 (Half)); + + /* Evaluate polynomial Q(x) = z + z * z2 * P(z2) with + z2 = x ^ 2 and z = |x| , if |x| < 0.5 + z2 = (1 - |x|) / 2 and z = sqrt(z2), if |x| >= 0.5. */ + float32x4_t z2 = vbslq_f32 (a_le_half, vmulq_f32 (x, x), + vfmsq_n_f32 (v_f32 (0.5), ax, 0.5)); + float32x4_t z = vbslq_f32 (a_le_half, ax, vsqrtq_f32 (z2)); + + /* Use a single polynomial approximation P for both intervals. */ + float32x4_t p = v_horner_4_f32 (z2, d->poly); + /* Finalize polynomial: z + z * z2 * P(z2). */ + p = vfmaq_f32 (z, vmulq_f32 (z, z2), p); + + /* acos(|x|) = pi/2 - sign(x) * Q(|x|), for |x| < 0.5 + = 2 Q(|x|) , for 0.5 < x < 1.0 + = pi - 2 Q(|x|) , for -1.0 < x < -0.5. */ + float32x4_t y = vbslq_f32 (v_u32 (AbsMask), p, x); + + uint32x4_t is_neg = vcltzq_f32 (x); + float32x4_t off = vreinterpretq_f32_u32 ( + vandq_u32 (vreinterpretq_u32_f32 (d->pif), is_neg)); + float32x4_t mul = vbslq_f32 (a_le_half, v_f32 (-1.0), v_f32 (2.0)); + float32x4_t add = vbslq_f32 (a_le_half, d->pi_over_2f, off); + + return vfmaq_f32 (add, mul, y); +} diff --git a/sysdeps/aarch64/fpu/acosf_sve.c b/sysdeps/aarch64/fpu/acosf_sve.c new file mode 100644 index 0000000000..44253fa999 --- /dev/null +++ b/sysdeps/aarch64/fpu/acosf_sve.c @@ -0,0 +1,86 @@ +/* Single-precision SVE inverse cos + + Copyright (C) 2023 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + . */ + +#include "sv_math.h" +#include "poly_sve_f32.h" + +static const struct data +{ + float32_t poly[5]; + float32_t pi, pi_over_2; +} data = { + /* Polynomial approximation of (asin(sqrt(x)) - sqrt(x)) / (x * sqrt(x)) on + [ 0x1p-24 0x1p-2 ] order = 4 rel error: 0x1.00a23bbp-29 . */ + .poly = { 0x1.55555ep-3, 0x1.33261ap-4, 0x1.70d7dcp-5, 0x1.b059dp-6, + 0x1.3af7d8p-5, }, + .pi = 0x1.921fb6p+1f, + .pi_over_2 = 0x1.921fb6p+0f, +}; + +/* Single-precision SVE implementation of vector acos(x). + + For |x| in [0, 0.5], use order 4 polynomial P such that the final + approximation of asin is an odd polynomial: + + acos(x) ~ pi/2 - (x + x^3 P(x^2)). + + The largest observed error in this region is 1.16 ulps, + _ZGVsMxv_acosf(0x1.ffbeccp-2) got 0x1.0c27f8p+0 + want 0x1.0c27f6p+0. + + For |x| in [0.5, 1.0], use same approximation with a change of variable + + acos(x) = y + y * z * P(z), with z = (1-x)/2 and y = sqrt(z). + + The largest observed error in this region is 1.32 ulps, + _ZGVsMxv_acosf (0x1.15ba56p-1) got 0x1.feb33p-1 + want 0x1.feb32ep-1. */ +svfloat32_t SV_NAME_F1 (acos) (svfloat32_t x, const svbool_t pg) +{ + const struct data *d = ptr_barrier (&data); + + svuint32_t sign = svand_x (pg, svreinterpret_u32 (x), 0x80000000); + svfloat32_t ax = svabs_x (pg, x); + svbool_t a_gt_half = svacgt (pg, x, 0.5); + + /* Evaluate polynomial Q(x) = z + z * z2 * P(z2) with + z2 = x ^ 2 and z = |x| , if |x| < 0.5 + z2 = (1 - |x|) / 2 and z = sqrt(z2), if |x| >= 0.5. */ + svfloat32_t z2 = svsel (a_gt_half, svmls_x (pg, sv_f32 (0.5), ax, 0.5), + svmul_x (pg, x, x)); + svfloat32_t z = svsqrt_m (ax, a_gt_half, z2); + + /* Use a single polynomial approximation P for both intervals. */ + svfloat32_t p = sv_horner_4_f32_x (pg, z2, d->poly); + /* Finalize polynomial: z + z * z2 * P(z2). */ + p = svmla_x (pg, z, svmul_x (pg, z, z2), p); + + /* acos(|x|) = pi/2 - sign(x) * Q(|x|), for |x| < 0.5 + = 2 Q(|x|) , for 0.5 < x < 1.0 + = pi - 2 Q(|x|) , for -1.0 < x < -0.5. */ + svfloat32_t y + = svreinterpret_f32 (svorr_x (pg, svreinterpret_u32 (p), sign)); + + svbool_t is_neg = svcmplt (pg, x, 0.0); + svfloat32_t off = svdup_f32_z (is_neg, d->pi); + svfloat32_t mul = svsel (a_gt_half, sv_f32 (2.0), sv_f32 (-1.0)); + svfloat32_t add = svsel (a_gt_half, off, sv_f32 (d->pi_over_2)); + + return svmla_x (pg, add, mul, y); +} diff --git a/sysdeps/aarch64/fpu/bits/math-vector.h b/sysdeps/aarch64/fpu/bits/math-vector.h index 03778faf96..f313993d70 100644 --- a/sysdeps/aarch64/fpu/bits/math-vector.h +++ b/sysdeps/aarch64/fpu/bits/math-vector.h @@ -49,6 +49,7 @@ typedef __SVBool_t __sv_bool_t; # define __vpcs __attribute__ ((__aarch64_vector_pcs__)) +__vpcs __f32x4_t _ZGVnN4v_acosf (__f32x4_t); __vpcs __f32x4_t _ZGVnN4v_asinf (__f32x4_t); __vpcs __f32x4_t _ZGVnN4v_cosf (__f32x4_t); __vpcs __f32x4_t _ZGVnN4v_expf (__f32x4_t); @@ -60,6 +61,7 @@ __vpcs __f32x4_t _ZGVnN4v_log2f (__f32x4_t); __vpcs __f32x4_t _ZGVnN4v_sinf (__f32x4_t); __vpcs __f32x4_t _ZGVnN4v_tanf (__f32x4_t); +__vpcs __f64x2_t _ZGVnN2v_acos (__f64x2_t); __vpcs __f64x2_t _ZGVnN2v_asin (__f64x2_t); __vpcs __f64x2_t _ZGVnN2v_cos (__f64x2_t); __vpcs __f64x2_t _ZGVnN2v_exp (__f64x2_t); @@ -76,6 +78,7 @@ __vpcs __f64x2_t _ZGVnN2v_tan (__f64x2_t); #ifdef __SVE_VEC_MATH_SUPPORTED +__sv_f32_t _ZGVsMxv_acosf (__sv_f32_t, __sv_bool_t); __sv_f32_t _ZGVsMxv_asinf (__sv_f32_t, __sv_bool_t); __sv_f32_t _ZGVsMxv_cosf (__sv_f32_t, __sv_bool_t); __sv_f32_t _ZGVsMxv_expf (__sv_f32_t, __sv_bool_t); @@ -87,6 +90,7 @@ __sv_f32_t _ZGVsMxv_log2f (__sv_f32_t, __sv_bool_t); __sv_f32_t _ZGVsMxv_sinf (__sv_f32_t, __sv_bool_t); __sv_f32_t _ZGVsMxv_tanf (__sv_f32_t, __sv_bool_t); +__sv_f64_t _ZGVsMxv_acos (__sv_f64_t, __sv_bool_t); __sv_f64_t _ZGVsMxv_asin (__sv_f64_t, __sv_bool_t); __sv_f64_t _ZGVsMxv_cos (__sv_f64_t, __sv_bool_t); __sv_f64_t _ZGVsMxv_exp (__sv_f64_t, __sv_bool_t); diff --git a/sysdeps/aarch64/fpu/test-double-advsimd-wrappers.c b/sysdeps/aarch64/fpu/test-double-advsimd-wrappers.c index b5ccd6b1cc..5a0cbf743b 100644 --- a/sysdeps/aarch64/fpu/test-double-advsimd-wrappers.c +++ b/sysdeps/aarch64/fpu/test-double-advsimd-wrappers.c @@ -23,6 +23,7 @@ #define VEC_TYPE float64x2_t +VPCS_VECTOR_WRAPPER (acos_advsimd, _ZGVnN2v_acos) VPCS_VECTOR_WRAPPER (asin_advsimd, _ZGVnN2v_asin) VPCS_VECTOR_WRAPPER (cos_advsimd, _ZGVnN2v_cos) VPCS_VECTOR_WRAPPER (exp_advsimd, _ZGVnN2v_exp) diff --git a/sysdeps/aarch64/fpu/test-double-sve-wrappers.c b/sysdeps/aarch64/fpu/test-double-sve-wrappers.c index fc3b20f421..bd89ff6133 100644 --- a/sysdeps/aarch64/fpu/test-double-sve-wrappers.c +++ b/sysdeps/aarch64/fpu/test-double-sve-wrappers.c @@ -32,6 +32,7 @@ return svlastb_f64 (svptrue_b64 (), mr); \ } +SVE_VECTOR_WRAPPER (acos_sve, _ZGVsMxv_acos) SVE_VECTOR_WRAPPER (asin_sve, _ZGVsMxv_asin) SVE_VECTOR_WRAPPER (cos_sve, _ZGVsMxv_cos) SVE_VECTOR_WRAPPER (exp_sve, _ZGVsMxv_exp) diff --git a/sysdeps/aarch64/fpu/test-float-advsimd-wrappers.c b/sysdeps/aarch64/fpu/test-float-advsimd-wrappers.c index 0a36aa91f5..3fafca7557 100644 --- a/sysdeps/aarch64/fpu/test-float-advsimd-wrappers.c +++ b/sysdeps/aarch64/fpu/test-float-advsimd-wrappers.c @@ -23,6 +23,7 @@ #define VEC_TYPE float32x4_t +VPCS_VECTOR_WRAPPER (acosf_advsimd, _ZGVnN4v_acosf) VPCS_VECTOR_WRAPPER (asinf_advsimd, _ZGVnN4v_asinf) VPCS_VECTOR_WRAPPER (cosf_advsimd, _ZGVnN4v_cosf) VPCS_VECTOR_WRAPPER (expf_advsimd, _ZGVnN4v_expf) diff --git a/sysdeps/aarch64/fpu/test-float-sve-wrappers.c b/sysdeps/aarch64/fpu/test-float-sve-wrappers.c index f7e4882c7a..b4ec9f777b 100644 --- a/sysdeps/aarch64/fpu/test-float-sve-wrappers.c +++ b/sysdeps/aarch64/fpu/test-float-sve-wrappers.c @@ -32,6 +32,7 @@ return svlastb_f32 (svptrue_b32 (), mr); \ } +SVE_VECTOR_WRAPPER (acosf_sve, _ZGVsMxv_acosf) SVE_VECTOR_WRAPPER (asinf_sve, _ZGVsMxv_asinf) SVE_VECTOR_WRAPPER (cosf_sve, _ZGVsMxv_cosf) SVE_VECTOR_WRAPPER (expf_sve, _ZGVsMxv_expf) diff --git a/sysdeps/aarch64/libm-test-ulps b/sysdeps/aarch64/libm-test-ulps index 1edc0fc343..c2b6f21b9d 100644 --- a/sysdeps/aarch64/libm-test-ulps +++ b/sysdeps/aarch64/libm-test-ulps @@ -6,11 +6,19 @@ double: 1 float: 1 ldouble: 1 +Function: "acos_advsimd": +double: 1 +float: 1 + Function: "acos_downward": double: 1 float: 1 ldouble: 1 +Function: "acos_sve": +double: 1 +float: 1 + Function: "acos_towardzero": double: 1 float: 1 diff --git a/sysdeps/unix/sysv/linux/aarch64/libmvec.abilist b/sysdeps/unix/sysv/linux/aarch64/libmvec.abilist index 6431c3fe65..f79eaaf241 100644 --- a/sysdeps/unix/sysv/linux/aarch64/libmvec.abilist +++ b/sysdeps/unix/sysv/linux/aarch64/libmvec.abilist @@ -14,18 +14,22 @@ GLIBC_2.38 _ZGVsMxv_log F GLIBC_2.38 _ZGVsMxv_logf F GLIBC_2.38 _ZGVsMxv_sin F GLIBC_2.38 _ZGVsMxv_sinf F +GLIBC_2.39 _ZGVnN2v_acos F GLIBC_2.39 _ZGVnN2v_asin F GLIBC_2.39 _ZGVnN2v_exp10 F GLIBC_2.39 _ZGVnN2v_exp2 F GLIBC_2.39 _ZGVnN2v_log10 F GLIBC_2.39 _ZGVnN2v_log2 F GLIBC_2.39 _ZGVnN2v_tan F +GLIBC_2.39 _ZGVnN4v_acosf F GLIBC_2.39 _ZGVnN4v_asinf F GLIBC_2.39 _ZGVnN4v_exp10f F GLIBC_2.39 _ZGVnN4v_exp2f F GLIBC_2.39 _ZGVnN4v_log10f F GLIBC_2.39 _ZGVnN4v_log2f F GLIBC_2.39 _ZGVnN4v_tanf F +GLIBC_2.39 _ZGVsMxv_acos F +GLIBC_2.39 _ZGVsMxv_acosf F GLIBC_2.39 _ZGVsMxv_asin F GLIBC_2.39 _ZGVsMxv_asinf F GLIBC_2.39 _ZGVsMxv_exp10 F