From patchwork Tue Nov 5 18:30:39 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Haris Okanovic X-Patchwork-Id: 13863436 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 439CCD328A4 for ; Tue, 5 Nov 2024 18:38:37 +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=jU1idzmOqOXlXb1t2IMpZrPTuLA9eFztDBmS3LLVVOw=; b=kbkanbIwqKRScOgI8A8zy8piKB /VaRBmEzUURDf5kR7wEj3KmQXfbMsVhPtJ3N0dI7EKMz1+LxBq87p2f1zq0VFWJeLGBTu0ZQ7s9Fv 78TFHfFXRa3+K/qrIoc5uDLmlz/Nz5RzUCc+NSDXqGBObFAcPV29gUnsVSUgktlkQ+23GB0vgdUDf EiaB56a5RJ5hiqqCmnnJ7KcrEBZKhbLFukhlc3KByYBHZjPBqtAz66pfSyBiC2VZ/IgtaWBR+iAzM oyFAAwHGGj5emJ/HCYNOyWcbKScfryH1panV44peyO0BGxqyIWBil24WRgpgmHc/lcC6yB6hFf9v+ ZnDKsP6A==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.98 #2 (Red Hat Linux)) id 1t8ORd-00000000Oic-1A3e; Tue, 05 Nov 2024 18:38:25 +0000 Received: from smtp-fw-9106.amazon.com ([207.171.188.206]) by bombadil.infradead.org with esmtps (Exim 4.98 #2 (Red Hat Linux)) id 1t8OLg-00000000NYm-1DDP for linux-arm-kernel@lists.infradead.org; Tue, 05 Nov 2024 18:32:18 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=amazon.com; i=@amazon.com; q=dns/txt; s=amazon201209; t=1730831536; x=1762367536; h=from:to:cc:subject:date:message-id:in-reply-to: references:mime-version:content-transfer-encoding; bh=jU1idzmOqOXlXb1t2IMpZrPTuLA9eFztDBmS3LLVVOw=; b=W2WNybKHNiaqe0fI03wsDQSkjpM3gn76FGWkycZNpaJA5eLt1UGteJvp R2t9t6rbgZFX+PkaKSvXOIIuGdI/D3A0BdS4DhwExTS7JSfUBNBwXjTL6 ntNwKxS0DzzSG/AuaQVuIQlvlvYkXZk/bwuRxww1a+pMKzUzyLcVO0qLB Y=; X-IronPort-AV: E=Sophos;i="6.11,260,1725321600"; d="scan'208";a="773030717" Received: from pdx4-co-svc-p1-lb2-vlan2.amazon.com (HELO smtpout.prod.us-west-2.prod.farcaster.email.amazon.dev) ([10.25.36.210]) by smtp-border-fw-9106.sea19.amazon.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 05 Nov 2024 18:32:13 +0000 Received: from EX19MTAUWA001.ant.amazon.com [10.0.21.151:34625] by smtpin.naws.us-west-2.prod.farcaster.email.amazon.dev [10.0.28.125:2525] with esmtp (Farcaster) id f9c0b0d1-8be1-4e25-ab5f-c68efeaab89c; Tue, 5 Nov 2024 18:32:12 +0000 (UTC) X-Farcaster-Flow-ID: f9c0b0d1-8be1-4e25-ab5f-c68efeaab89c Received: from EX19D001UWA003.ant.amazon.com (10.13.138.211) by EX19MTAUWA001.ant.amazon.com (10.250.64.217) 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 18:32:12 +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 18:32:06 +0000 From: Haris Okanovic To: , CC: , , , , , , , , , , , , , , , , , , , , , , , , , , , , Subject: [PATCH 3/5] arm64: refactor delay() to enable polling for value Date: Tue, 5 Nov 2024 12:30:39 -0600 Message-ID: <20241105183041.1531976-4-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_103216_486432_B6602B4A X-CRM114-Status: GOOD ( 13.28 ) 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 Refactor arm64's delay() to poll for a mask/value condition (vcond) in it's wfet(), wfe(), and relaxed polling loops. Signed-off-by: Haris Okanovic --- arch/arm64/lib/delay.c | 70 ++++++++++++++++++++++++++++++------------ 1 file changed, 50 insertions(+), 20 deletions(-) diff --git a/arch/arm64/lib/delay.c b/arch/arm64/lib/delay.c index cb2062e7e234..a7c3040af316 100644 --- a/arch/arm64/lib/delay.c +++ b/arch/arm64/lib/delay.c @@ -14,43 +14,73 @@ #include #include +#include -#define USECS_TO_CYCLES(time_usecs) \ - xloops_to_cycles((time_usecs) * 0x10C7UL) - -static inline unsigned long xloops_to_cycles(unsigned long xloops) +static inline u64 xloops_to_cycles(u64 xloops) { return (xloops * loops_per_jiffy * HZ) >> 32; } -void __delay(unsigned long cycles) +#define USECS_TO_XLOOPS(time_usecs) \ + ((time_usecs) * 0x10C7UL) + +#define USECS_TO_CYCLES(time_usecs) \ + xloops_to_cycles(USECS_TO_XLOOPS(time_usecs)) + +#define NSECS_TO_XLOOPS(time_nsecs) \ + ((time_nsecs) * 0x10C7UL) + +#define NSECS_TO_CYCLES(time_nsecs) \ + xloops_to_cycles(NSECS_TO_XLOOPS(time_nsecs)) + +static unsigned long __delay_until_ul(u64 cycles, unsigned long* addr, unsigned long mask, unsigned long val) { - cycles_t start = get_cycles(); + u64 start = get_cycles(); + unsigned long cur; if (alternative_has_cap_unlikely(ARM64_HAS_WFXT)) { u64 end = start + cycles; - /* - * Start with WFIT. If an interrupt makes us resume - * early, use a WFET loop to complete the delay. - */ - wfit(end); - while ((get_cycles() - start) < cycles) + do { + cur = __READ_ONCE_EX(*addr); + if ((cur & mask) == val) { + break; + } wfet(end); - } else if (arch_timer_evtstrm_available()) { - const cycles_t timer_evt_period = + } while ((get_cycles() - start) < cycles); + } else if (arch_timer_evtstrm_available()) { + const u64 timer_evt_period = USECS_TO_CYCLES(ARCH_TIMER_EVT_STREAM_PERIOD_US); - while ((get_cycles() - start + timer_evt_period) < cycles) + do { + cur = __READ_ONCE_EX(*addr); + if ((cur & mask) == val) { + break; + } wfe(); + } while ((get_cycles() - start + timer_evt_period) < cycles); + } else { + do { + cur = __READ_ONCE_EX(*addr); + if ((cur & mask) == val) { + break; + } + cpu_relax(); + } while ((get_cycles() - start) < cycles); } - while ((get_cycles() - start) < cycles) - cpu_relax(); + return cur; +} + +void __delay(unsigned long cycles) +{ + /* constant word for wfet()/wfe() to poll */ + unsigned long dummy ____cacheline_aligned = 0; + __delay_until_ul(cycles, &dummy, 0, 1); } EXPORT_SYMBOL(__delay); -inline void __const_udelay(unsigned long xloops) +void __const_udelay(unsigned long xloops) { __delay(xloops_to_cycles(xloops)); } @@ -58,12 +88,12 @@ EXPORT_SYMBOL(__const_udelay); void __udelay(unsigned long usecs) { - __const_udelay(usecs * 0x10C7UL); /* 2**32 / 1000000 (rounded up) */ + __delay(USECS_TO_CYCLES(usecs)); } EXPORT_SYMBOL(__udelay); void __ndelay(unsigned long nsecs) { - __const_udelay(nsecs * 0x5UL); /* 2**32 / 1000000000 (rounded up) */ + __delay(NSECS_TO_CYCLES(nsecs)); } EXPORT_SYMBOL(__ndelay);