From patchwork Tue Nov 5 18:30:40 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Okanovic, Haris" X-Patchwork-Id: 13863437 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 01F72D33A20 for ; Tue, 5 Nov 2024 18:40:21 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:List-Subscribe:List-Help :List-Post:List-Archive:List-Unsubscribe:List-Id:Content-Type: Content-Transfer-Encoding:MIME-Version:References:In-Reply-To:Message-ID:Date :Subject:CC:To:From:Reply-To:Content-ID:Content-Description:Resent-Date: Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID:List-Owner; bh=XT81pCaCySeLSnjYgvTXLKqn2CWQcbY+GGEiD2145i8=; b=WYpYSmRErQj3WmSjnU+EzZxABy oBS+RfplcVrGUn0iVgZPEGaakER7cVU+5RztmXJniX8B1MLfr8bQeeFGFrWU67gz+XeRm4zdV4z7m PWy3i/WD81KdDVu53wCdsNGLeKqzSB/PBeyEZgLemenNU33YlYS8fwUCYVFHe7JiId4fLtV83yA2Z gDGW/fSWcarTxiaKhBnzqT5/kgF6eqn2UlA94ILFLdvAxTTRQhbhbq9NjaB7QJV1wUETbzupzAzFz 1a6ALl9etWDasQOhBmZN/9KKvIc+q9oxJejUflwU2+YEAL90msyXQR1EktS44yiLYodq0hD+KPyCC jTNWhdJQ==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.98 #2 (Red Hat Linux)) id 1t8OTM-00000000PFI-2EhZ; Tue, 05 Nov 2024 18:40:12 +0000 Received: from smtp-fw-52003.amazon.com ([52.119.213.152]) by bombadil.infradead.org with esmtps (Exim 4.98 #2 (Red Hat Linux)) id 1t8OLp-00000000NcS-2pqo for linux-arm-kernel@lists.infradead.org; Tue, 05 Nov 2024 18:32:27 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=amazon.com; i=@amazon.com; q=dns/txt; s=amazon201209; t=1730831546; x=1762367546; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=XT81pCaCySeLSnjYgvTXLKqn2CWQcbY+GGEiD2145i8=; b=qm8lqGbr00hwgViKL07yykqvVvS7Is6gVgyheDXiVaQ+t/8Pu397Dfq8 UxuZ3kOQDrQGthtJEFFdPh/GTkTBmZQFzUHyrWbJnFDqNsFuoHagn9RmX 7gNrl63jpZJjeQzM0rv0AsRHHqX+RN/a84mzobuRhGz+/s/FuHaU+Pg1d M=; X-IronPort-AV: E=Sophos;i="6.11,260,1725321600"; d="scan'208";a="39203526" Received: from iad12-co-svc-p1-lb1-vlan3.amazon.com (HELO smtpout.prod.us-west-2.prod.farcaster.email.amazon.dev) ([10.43.8.6]) by smtp-border-fw-52003.iad7.amazon.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Nov 2024 18:32:22 +0000 Received: from EX19MTAUWC001.ant.amazon.com [10.0.7.35:25778] by smtpin.naws.us-west-2.prod.farcaster.email.amazon.dev [10.0.63.130:2525] with esmtp (Farcaster) id fb722a64-c7d9-4601-a9fb-a5752e9890eb; Tue, 5 Nov 2024 18:32:20 +0000 (UTC) X-Farcaster-Flow-ID: fb722a64-c7d9-4601-a9fb-a5752e9890eb Received: from EX19D001UWA003.ant.amazon.com (10.13.138.211) by EX19MTAUWC001.ant.amazon.com (10.250.64.174) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA) id 15.2.1258.34; Tue, 5 Nov 2024 18:32:20 +0000 Received: from u34cccd802f2d52.amazon.com (10.106.239.17) by EX19D001UWA003.ant.amazon.com (10.13.138.211) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA) id 15.2.1258.35; Tue, 5 Nov 2024 18:32:15 +0000 From: Haris Okanovic To: , CC: , , , , , , , , , , , , , , , , , , , , , , , , , , , , Subject: [PATCH 4/5] arm64: add smp_vcond_load_relaxed() Date: Tue, 5 Nov 2024 12:30:40 -0600 Message-ID: <20241105183041.1531976-5-harisokn@amazon.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20241105183041.1531976-1-harisokn@amazon.com> References: <20240925232425.2763385-1-ankur.a.arora@oracle.com> <20241105183041.1531976-1-harisokn@amazon.com> MIME-Version: 1.0 X-Originating-IP: [10.106.239.17] X-ClientProxiedBy: EX19D033UWA004.ant.amazon.com (10.13.139.85) To EX19D001UWA003.ant.amazon.com (10.13.138.211) X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20241105_103225_904291_E56389D2 X-CRM114-Status: GOOD ( 10.54 ) X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=archiver.kernel.org@lists.infradead.org Implement smp_vcond_load_relaxed() atop __delay_until_ul() on arm64, to reduce number of busy loops while waiting for a value condition. This implementation only support unsigned long words. It can be extended via the enclosed case structure in barrier.h as needed. Signed-off-by: Haris Okanovic --- arch/arm64/include/asm/barrier.h | 18 ++++++++++++++++++ arch/arm64/lib/delay.c | 16 ++++++++++++++++ 2 files changed, 34 insertions(+) diff --git a/arch/arm64/include/asm/barrier.h b/arch/arm64/include/asm/barrier.h index 1ca947d5c939..188327e3ce72 100644 --- a/arch/arm64/include/asm/barrier.h +++ b/arch/arm64/include/asm/barrier.h @@ -203,6 +203,24 @@ do { \ (typeof(*ptr))VAL; \ }) +extern unsigned long __smp_vcond_load_relaxed_ul( + u64 nsecs, unsigned long* addr, unsigned long mask, unsigned long val); + +#define smp_vcond_load_relaxed(nsecs, addr, mask, val) ({ \ + u64 __nsecs = (nsecs); \ + typeof(addr) __addr = (addr); \ + typeof(*__addr) __mask = (mask); \ + typeof(*__addr) __val = (val); \ + typeof(*__addr) __cur; \ + switch (sizeof(*__addr)) { \ + case sizeof(unsigned long): \ + __cur = __smp_vcond_load_relaxed_ul( \ + __nsecs, __addr, __mask, __val); \ + break; \ + } \ + (__cur); \ +}) + #define smp_cond_load_acquire(ptr, cond_expr) \ ({ \ typeof(ptr) __PTR = (ptr); \ diff --git a/arch/arm64/lib/delay.c b/arch/arm64/lib/delay.c index a7c3040af316..a61a13b04439 100644 --- a/arch/arm64/lib/delay.c +++ b/arch/arm64/lib/delay.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include @@ -97,3 +98,18 @@ void __ndelay(unsigned long nsecs) __delay(NSECS_TO_CYCLES(nsecs)); } EXPORT_SYMBOL(__ndelay); + +unsigned long __smp_vcond_load_relaxed_ul( + u64 nsecs, unsigned long* addr, unsigned long mask, unsigned long val) +{ + const u64 start = local_clock_noinstr(); + const u64 cycles = NSECS_TO_CYCLES(nsecs); + unsigned long cur; + + do { + cur = __delay_until_ul(cycles, addr, mask, val); + } while((cur & mask) != val && local_clock_noinstr() - start < nsecs); + + return cur; +} +EXPORT_SYMBOL(__smp_vcond_load_relaxed_ul);