From nobody Fri Dec 19 16:53:31 2025 Received: from mx0b-0016f401.pphosted.com (mx0a-0016f401.pphosted.com [67.231.148.174]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id ACD8C1E1A04 for ; Wed, 4 Sep 2024 18:26:38 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=67.231.148.174 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474400; cv=none; b=Uavo8CQ70cpfl9GShw8Xhyb6XLKPgdlZmTSzv1o7MGLyD+kUvo0eYjActb4O9F2l+Wuaq03WE+Nc5RJZlK5feSJNeU6UNYYOyhuEphyQIx9q7BXwAQ1RRkPYlvnPOcidfavriDnAbnhO716Af3KskVbfPqW5BX2QNHf31gNvPNg= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474400; c=relaxed/simple; bh=Aoh/8y5Lknrp0RPwDNnrveLn3DXPHCRXdW/8xcrokoE=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=I9hub2Cbvoa/CeULM9U5jaVJVBc6oGPQw6rkh63iKB3B8ldVQfdPJ1pHnCQpHaMQKFFJvEjSpxb4HioLhfv8Uhu2dIGWaa+iPetM9DfK6h8u/+nRtWX83xfAh7aJ0ssHcspc3K/YkgJba97j/pObc3yPkEs2Cbnqm28bpBmHO0E= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com; spf=pass smtp.mailfrom=marvell.com; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b=kF/j1hda; arc=none smtp.client-ip=67.231.148.174 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=marvell.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b="kF/j1hda" Received: from pps.filterd (m0045849.ppops.net [127.0.0.1]) by mx0a-0016f401.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 484EHHaI010434; Wed, 4 Sep 2024 11:26:30 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pfpt0220; bh=p vcnGVyI8Dw5pDCMMbupUgw+CsZJKDfzxSi9l4dxRlc=; b=kF/j1hdaXG/R/XWIp AGU/IPJMu5hbrK9U43NKjkcpHNlDWqs/7PfAGVDRzShR8ccys5ZoF9Ox5mUZnx0W 1nq8cbx+hDN5QZSNjpVa5xC4JlImbpKZFmTza/k5NweFsrSAdw+3J5x9h93yzMGO ndbDPKgjvUumK6duuHA9KFgJV1mVZpdEkua+0zZCCjP9jnmv/wyN4NwnEykMZbHU cgzAL+i/x0AGaHBuKMVQ5SajnHI9aANh9goI0nSKMcTy/1I8oI6l8x4d+WBpFLMq nvfhPyRHhn4/xJRxLj4JsbL4AKlDYV+aWd/N0Z3a9b37x4fDnEbdaYPPro0p95RJ CAyog== Received: from dc6wp-exch02.marvell.com ([4.21.29.225]) by mx0a-0016f401.pphosted.com (PPS) with ESMTPS id 41es66h578-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 04 Sep 2024 11:26:30 -0700 (PDT) Received: from DC6WP-EXCH02.marvell.com (10.76.176.209) by DC6WP-EXCH02.marvell.com (10.76.176.209) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.4; Wed, 4 Sep 2024 11:26:28 -0700 Received: from maili.marvell.com (10.69.176.80) by DC6WP-EXCH02.marvell.com (10.76.176.209) with Microsoft SMTP Server id 15.2.1544.4 via Frontend Transport; Wed, 4 Sep 2024 11:26:28 -0700 Received: from IPBU-BLR-SERVER1.marvell.com (IPBU-BLR-SERVER1.marvell.com [10.28.8.41]) by maili.marvell.com (Postfix) with ESMTP id 059703F7044; Wed, 4 Sep 2024 11:26:25 -0700 (PDT) From: Gowthami Thiagarajan To: , , , CC: , , , Gowthami Thiagarajan Subject: [PATCH v7 1/6] perf/marvell: Refactor to extract platform data - no functional change Date: Wed, 4 Sep 2024 23:56:00 +0530 Message-ID: <20240904182605.953927-2-gthiagarajan@marvell.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240904182605.953927-1-gthiagarajan@marvell.com> References: <20240904182605.953927-1-gthiagarajan@marvell.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Proofpoint-GUID: CUDi10bky4PcP9bPPkOWO5Hs3MHDVyaR X-Proofpoint-ORIG-GUID: CUDi10bky4PcP9bPPkOWO5Hs3MHDVyaR X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-04_16,2024-09-04_01,2024-09-02_01 Content-Type: text/plain; charset="utf-8" This commit introduces a refactor to the Marvell DDR pmu driver, specifically targeting the extraction of platform data (referred to as "pdata") from the existing driver. The purpose of this refactor is to prepare for the upcoming support of the next version of the Performance Monitoring Unit (PMU) in this driver. No functional changes are introduced in this refactor. Its sole purpose is to improve code organization and pave the way for future enhancements to the driver. Signed-off-by: Gowthami Thiagarajan --- drivers/perf/marvell_cn10k_ddr_pmu.c | 127 +++++++++++++++++++-------- 1 file changed, 92 insertions(+), 35 deletions(-) diff --git a/drivers/perf/marvell_cn10k_ddr_pmu.c b/drivers/perf/marvell_cn= 10k_ddr_pmu.c index 94f1ebcd2a27..e33d383aa6d2 100644 --- a/drivers/perf/marvell_cn10k_ddr_pmu.c +++ b/drivers/perf/marvell_cn10k_ddr_pmu.c @@ -1,7 +1,8 @@ // SPDX-License-Identifier: GPL-2.0 -/* Marvell CN10K DRAM Subsystem (DSS) Performance Monitor Driver +/* + * Marvell CN10K DRAM Subsystem (DSS) Performance Monitor Driver * - * Copyright (C) 2021 Marvell. + * Copyright (C) 2024 Marvell. */ =20 #include @@ -14,24 +15,24 @@ #include =20 /* Performance Counters Operating Mode Control Registers */ -#define DDRC_PERF_CNT_OP_MODE_CTRL 0x8020 -#define OP_MODE_CTRL_VAL_MANNUAL 0x1 +#define CN10K_DDRC_PERF_CNT_OP_MODE_CTRL 0x8020 +#define OP_MODE_CTRL_VAL_MANUAL 0x1 =20 /* Performance Counters Start Operation Control Registers */ -#define DDRC_PERF_CNT_START_OP_CTRL 0x8028 +#define CN10K_DDRC_PERF_CNT_START_OP_CTRL 0x8028 #define START_OP_CTRL_VAL_START 0x1ULL #define START_OP_CTRL_VAL_ACTIVE 0x2 =20 /* Performance Counters End Operation Control Registers */ -#define DDRC_PERF_CNT_END_OP_CTRL 0x8030 +#define CN10K_DDRC_PERF_CNT_END_OP_CTRL 0x8030 #define END_OP_CTRL_VAL_END 0x1ULL =20 /* Performance Counters End Status Registers */ -#define DDRC_PERF_CNT_END_STATUS 0x8038 +#define CN10K_DDRC_PERF_CNT_END_STATUS 0x8038 #define END_STATUS_VAL_END_TIMER_MODE_END 0x1 =20 /* Performance Counters Configuration Registers */ -#define DDRC_PERF_CFG_BASE 0x8040 +#define CN10K_DDRC_PERF_CFG_BASE 0x8040 =20 /* 8 Generic event counter + 2 fixed event counters */ #define DDRC_PERF_NUM_GEN_COUNTERS 8 @@ -42,17 +43,19 @@ DDRC_PERF_NUM_FIX_COUNTERS) =20 /* Generic event counter registers */ -#define DDRC_PERF_CFG(n) (DDRC_PERF_CFG_BASE + 8 * (n)) +#define DDRC_PERF_CFG(base, n) ((base) + 8 * (n)) #define EVENT_ENABLE BIT_ULL(63) =20 /* Two dedicated event counters for DDR reads and writes */ #define EVENT_DDR_READS 101 #define EVENT_DDR_WRITES 100 =20 +#define DDRC_PERF_REG(base, n) ((base) + 8 * (n)) /* * programmable events IDs in programmable event counters. * DO NOT change these event-id numbers, they are used to * program event bitmap in h/w. + * */ #define EVENT_OP_IS_ZQLATCH 55 #define EVENT_OP_IS_ZQSTART 54 @@ -63,8 +66,8 @@ #define EVENT_VISIBLE_WIN_LIMIT_REACHED_RD 49 #define EVENT_BSM_STARVATION 48 #define EVENT_BSM_ALLOC 47 -#define EVENT_LPR_REQ_WITH_NOCREDIT 46 -#define EVENT_HPR_REQ_WITH_NOCREDIT 45 +#define EVENT_RETRY_FIFO_FULL_OR_LPR_REQ_NOCRED 46 +#define EVENT_DFI_OR_HPR_REQ_NOCRED 45 #define EVENT_OP_IS_ZQCS 44 #define EVENT_OP_IS_ZQCL 43 #define EVENT_OP_IS_LOAD_MODE 42 @@ -102,28 +105,29 @@ #define EVENT_HIF_RD_OR_WR 1 =20 /* Event counter value registers */ -#define DDRC_PERF_CNT_VALUE_BASE 0x8080 -#define DDRC_PERF_CNT_VALUE(n) (DDRC_PERF_CNT_VALUE_BASE + 8 * (n)) +#define CN10K_DDRC_PERF_CNT_VALUE_BASE 0x8080 =20 /* Fixed event counter enable/disable register */ -#define DDRC_PERF_CNT_FREERUN_EN 0x80C0 +#define CN10K_DDRC_PERF_CNT_FREERUN_EN 0x80C0 #define DDRC_PERF_FREERUN_WRITE_EN 0x1 #define DDRC_PERF_FREERUN_READ_EN 0x2 =20 /* Fixed event counter control register */ -#define DDRC_PERF_CNT_FREERUN_CTRL 0x80C8 +#define CN10K_DDRC_PERF_CNT_FREERUN_CTRL 0x80C8 #define DDRC_FREERUN_WRITE_CNT_CLR 0x1 #define DDRC_FREERUN_READ_CNT_CLR 0x2 =20 -/* Fixed event counter value register */ -#define DDRC_PERF_CNT_VALUE_WR_OP 0x80D0 -#define DDRC_PERF_CNT_VALUE_RD_OP 0x80D8 #define DDRC_PERF_CNT_VALUE_OVERFLOW BIT_ULL(48) #define DDRC_PERF_CNT_MAX_VALUE GENMASK_ULL(48, 0) =20 +/* Fixed event counter value register */ +#define CN10K_DDRC_PERF_CNT_VALUE_WR_OP 0x80D0 +#define CN10K_DDRC_PERF_CNT_VALUE_RD_OP 0x80D8 + struct cn10k_ddr_pmu { struct pmu pmu; void __iomem *base; + const struct ddr_pmu_platform_data *p_data; unsigned int cpu; struct device *dev; int active_events; @@ -134,6 +138,22 @@ struct cn10k_ddr_pmu { =20 #define to_cn10k_ddr_pmu(p) container_of(p, struct cn10k_ddr_pmu, pmu) =20 +struct ddr_pmu_platform_data { + u64 counter_overflow_val; + u64 counter_max_val; + u64 ddrc_perf_cnt_base; + u64 ddrc_perf_cfg_base; + u64 ddrc_perf_cnt_op_mode_ctrl; + u64 ddrc_perf_cnt_start_op_ctrl; + u64 ddrc_perf_cnt_end_op_ctrl; + u64 ddrc_perf_cnt_end_status; + u64 ddrc_perf_cnt_freerun_en; + u64 ddrc_perf_cnt_freerun_ctrl; + u64 ddrc_perf_cnt_freerun_clr; + u64 ddrc_perf_cnt_value_wr_op; + u64 ddrc_perf_cnt_value_rd_op; +}; + static ssize_t cn10k_ddr_pmu_event_show(struct device *dev, struct device_attribute *attr, char *page) @@ -189,9 +209,9 @@ static struct attribute *cn10k_ddr_perf_events_attrs[] = =3D { CN10K_DDR_PMU_EVENT_ATTR(ddr_zqcl, EVENT_OP_IS_ZQCL), CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_wr_access, EVENT_OP_IS_ZQCS), CN10K_DDR_PMU_EVENT_ATTR(ddr_hpr_req_with_nocredit, - EVENT_HPR_REQ_WITH_NOCREDIT), + EVENT_DFI_OR_HPR_REQ_NOCRED), CN10K_DDR_PMU_EVENT_ATTR(ddr_lpr_req_with_nocredit, - EVENT_LPR_REQ_WITH_NOCREDIT), + EVENT_RETRY_FIFO_FULL_OR_LPR_REQ_NOCRED), CN10K_DDR_PMU_EVENT_ATTR(ddr_bsm_alloc, EVENT_BSM_ALLOC), CN10K_DDR_PMU_EVENT_ATTR(ddr_bsm_starvation, EVENT_BSM_STARVATION), CN10K_DDR_PMU_EVENT_ATTR(ddr_win_limit_reached_rd, @@ -354,6 +374,7 @@ static int cn10k_ddr_perf_event_init(struct perf_event = *event) static void cn10k_ddr_perf_counter_enable(struct cn10k_ddr_pmu *pmu, int counter, bool enable) { + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; u32 reg; u64 val; =20 @@ -363,7 +384,7 @@ static void cn10k_ddr_perf_counter_enable(struct cn10k_= ddr_pmu *pmu, } =20 if (counter < DDRC_PERF_NUM_GEN_COUNTERS) { - reg =3D DDRC_PERF_CFG(counter); + reg =3D DDRC_PERF_CFG(p_data->ddrc_perf_cfg_base, counter); val =3D readq_relaxed(pmu->base + reg); =20 if (enable) @@ -373,7 +394,8 @@ static void cn10k_ddr_perf_counter_enable(struct cn10k_= ddr_pmu *pmu, =20 writeq_relaxed(val, pmu->base + reg); } else { - val =3D readq_relaxed(pmu->base + DDRC_PERF_CNT_FREERUN_EN); + val =3D readq_relaxed(pmu->base + + p_data->ddrc_perf_cnt_freerun_en); if (enable) { if (counter =3D=3D DDRC_PERF_READ_COUNTER_IDX) val |=3D DDRC_PERF_FREERUN_READ_EN; @@ -385,27 +407,33 @@ static void cn10k_ddr_perf_counter_enable(struct cn10= k_ddr_pmu *pmu, else val &=3D ~DDRC_PERF_FREERUN_WRITE_EN; } - writeq_relaxed(val, pmu->base + DDRC_PERF_CNT_FREERUN_EN); + writeq_relaxed(val, pmu->base + + p_data->ddrc_perf_cnt_freerun_en); } } =20 static u64 cn10k_ddr_perf_read_counter(struct cn10k_ddr_pmu *pmu, int coun= ter) { + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; u64 val; =20 if (counter =3D=3D DDRC_PERF_READ_COUNTER_IDX) - return readq_relaxed(pmu->base + DDRC_PERF_CNT_VALUE_RD_OP); + return readq_relaxed(pmu->base + + p_data->ddrc_perf_cnt_value_rd_op); =20 if (counter =3D=3D DDRC_PERF_WRITE_COUNTER_IDX) - return readq_relaxed(pmu->base + DDRC_PERF_CNT_VALUE_WR_OP); + return readq_relaxed(pmu->base + + p_data->ddrc_perf_cnt_value_wr_op); =20 - val =3D readq_relaxed(pmu->base + DDRC_PERF_CNT_VALUE(counter)); + val =3D readq_relaxed(pmu->base + + DDRC_PERF_REG(p_data->ddrc_perf_cnt_base, counter)); return val; } =20 static void cn10k_ddr_perf_event_update(struct perf_event *event) { struct cn10k_ddr_pmu *pmu =3D to_cn10k_ddr_pmu(event->pmu); + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; struct hw_perf_event *hwc =3D &event->hw; u64 prev_count, new_count, mask; =20 @@ -414,7 +442,7 @@ static void cn10k_ddr_perf_event_update(struct perf_eve= nt *event) new_count =3D cn10k_ddr_perf_read_counter(pmu, hwc->idx); } while (local64_xchg(&hwc->prev_count, new_count) !=3D prev_count); =20 - mask =3D DDRC_PERF_CNT_MAX_VALUE; + mask =3D p_data->counter_max_val; =20 local64_add((new_count - prev_count) & mask, &event->count); } @@ -435,6 +463,7 @@ static void cn10k_ddr_perf_event_start(struct perf_even= t *event, int flags) static int cn10k_ddr_perf_event_add(struct perf_event *event, int flags) { struct cn10k_ddr_pmu *pmu =3D to_cn10k_ddr_pmu(event->pmu); + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; struct hw_perf_event *hwc =3D &event->hw; u8 config =3D event->attr.config; int counter, ret; @@ -454,7 +483,7 @@ static int cn10k_ddr_perf_event_add(struct perf_event *= event, int flags) =20 if (counter < DDRC_PERF_NUM_GEN_COUNTERS) { /* Generic counters, configure event id */ - reg_offset =3D DDRC_PERF_CFG(counter); + reg_offset =3D DDRC_PERF_CFG(p_data->ddrc_perf_cfg_base, counter); ret =3D ddr_perf_get_event_bitmap(config, &val); if (ret) return ret; @@ -467,7 +496,8 @@ static int cn10k_ddr_perf_event_add(struct perf_event *= event, int flags) else val =3D DDRC_FREERUN_WRITE_CNT_CLR; =20 - writeq_relaxed(val, pmu->base + DDRC_PERF_CNT_FREERUN_CTRL); + writeq_relaxed(val, pmu->base + + p_data->ddrc_perf_cnt_freerun_ctrl); } =20 hwc->state |=3D PERF_HES_STOPPED; @@ -512,17 +542,19 @@ static void cn10k_ddr_perf_event_del(struct perf_even= t *event, int flags) static void cn10k_ddr_perf_pmu_enable(struct pmu *pmu) { struct cn10k_ddr_pmu *ddr_pmu =3D to_cn10k_ddr_pmu(pmu); + const struct ddr_pmu_platform_data *p_data =3D ddr_pmu->p_data; =20 writeq_relaxed(START_OP_CTRL_VAL_START, ddr_pmu->base + - DDRC_PERF_CNT_START_OP_CTRL); + p_data->ddrc_perf_cnt_start_op_ctrl); } =20 static void cn10k_ddr_perf_pmu_disable(struct pmu *pmu) { struct cn10k_ddr_pmu *ddr_pmu =3D to_cn10k_ddr_pmu(pmu); + const struct ddr_pmu_platform_data *p_data =3D ddr_pmu->p_data; =20 writeq_relaxed(END_OP_CTRL_VAL_END, ddr_pmu->base + - DDRC_PERF_CNT_END_OP_CTRL); + p_data->ddrc_perf_cnt_end_op_ctrl); } =20 static void cn10k_ddr_perf_event_update_all(struct cn10k_ddr_pmu *pmu) @@ -549,6 +581,7 @@ static void cn10k_ddr_perf_event_update_all(struct cn10= k_ddr_pmu *pmu) =20 static irqreturn_t cn10k_ddr_pmu_overflow_handler(struct cn10k_ddr_pmu *pm= u) { + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; struct perf_event *event; struct hw_perf_event *hwc; u64 prev_count, new_count; @@ -586,7 +619,7 @@ static irqreturn_t cn10k_ddr_pmu_overflow_handler(struc= t cn10k_ddr_pmu *pmu) continue; =20 value =3D cn10k_ddr_perf_read_counter(pmu, i); - if (value =3D=3D DDRC_PERF_CNT_MAX_VALUE) { + if (value =3D=3D p_data->counter_max_val) { pr_info("Counter-(%d) reached max value\n", i); cn10k_ddr_perf_event_update_all(pmu); cn10k_ddr_perf_pmu_disable(&pmu->pmu); @@ -629,8 +662,25 @@ static int cn10k_ddr_pmu_offline_cpu(unsigned int cpu,= struct hlist_node *node) return 0; } =20 +static const struct ddr_pmu_platform_data cn10k_ddr_pmu_pdata =3D { + .counter_overflow_val =3D BIT_ULL(48), + .counter_max_val =3D GENMASK_ULL(48, 0), + .ddrc_perf_cnt_base =3D CN10K_DDRC_PERF_CNT_VALUE_BASE, + .ddrc_perf_cfg_base =3D CN10K_DDRC_PERF_CFG_BASE, + .ddrc_perf_cnt_op_mode_ctrl =3D CN10K_DDRC_PERF_CNT_OP_MODE_CTRL, + .ddrc_perf_cnt_start_op_ctrl =3D CN10K_DDRC_PERF_CNT_START_OP_CTRL, + .ddrc_perf_cnt_end_op_ctrl =3D CN10K_DDRC_PERF_CNT_END_OP_CTRL, + .ddrc_perf_cnt_end_status =3D CN10K_DDRC_PERF_CNT_END_STATUS, + .ddrc_perf_cnt_freerun_en =3D CN10K_DDRC_PERF_CNT_FREERUN_EN, + .ddrc_perf_cnt_freerun_ctrl =3D CN10K_DDRC_PERF_CNT_FREERUN_CTRL, + .ddrc_perf_cnt_freerun_clr =3D 0, + .ddrc_perf_cnt_value_wr_op =3D CN10K_DDRC_PERF_CNT_VALUE_WR_OP, + .ddrc_perf_cnt_value_rd_op =3D CN10K_DDRC_PERF_CNT_VALUE_RD_OP, +}; + static int cn10k_ddr_perf_probe(struct platform_device *pdev) { + const struct ddr_pmu_data *dev_data; struct cn10k_ddr_pmu *ddr_pmu; struct resource *res; void __iomem *base; @@ -644,15 +694,22 @@ static int cn10k_ddr_perf_probe(struct platform_devic= e *pdev) ddr_pmu->dev =3D &pdev->dev; platform_set_drvdata(pdev, ddr_pmu); =20 + dev_data =3D device_get_match_data(&pdev->dev); + if (!dev_data) { + dev_err(&pdev->dev, "Error: No device match data found\n"); + return -ENODEV; + } + base =3D devm_platform_get_and_ioremap_resource(pdev, 0, &res); if (IS_ERR(base)) return PTR_ERR(base); =20 ddr_pmu->base =3D base; =20 + ddr_pmu->p_data =3D &cn10k_ddr_pmu_pdata; /* Setup the PMU counter to work in manual mode */ - writeq_relaxed(OP_MODE_CTRL_VAL_MANNUAL, ddr_pmu->base + - DDRC_PERF_CNT_OP_MODE_CTRL); + writeq_relaxed(OP_MODE_CTRL_VAL_MANUAL, ddr_pmu->base + + ddr_pmu->p_data->ddrc_perf_cnt_op_mode_ctrl); =20 ddr_pmu->pmu =3D (struct pmu) { .module =3D THIS_MODULE, @@ -688,7 +745,7 @@ static int cn10k_ddr_perf_probe(struct platform_device = *pdev) if (ret) goto error; =20 - pr_info("CN10K DDR PMU Driver for ddrc@%llx\n", res->start); + pr_info("DDR PMU Driver for ddrc@%llx\n", res->start); return 0; error: cpuhp_state_remove_instance_nocalls( --=20 2.25.1 From nobody Fri Dec 19 16:53:31 2025 Received: from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com [67.231.156.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 6ADB81E4128 for ; Wed, 4 Sep 2024 18:26:44 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=67.231.156.173 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474406; cv=none; b=NG7uuzv149U18toby0Nh8BwaqBBwFeye7R2ngebNWkF3AVhpX6X0i4Pi0j6cFCHxnVVh60wp3JA6kfoZx+7U7idEl7PhmKbhE3sqajlswQyTKNs6v6mR4f0AJzJNQrOicCqyB15fZjZGBAuP+CDNoPc9vBKTxBY+hTS/yLfvrFo= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474406; c=relaxed/simple; bh=rRk741S3MwZtXCXG6ZQJgpfyrfk1ISD1p0Hml4Jv2y8=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=s3myz1nbPJkUrmEsS0QWUxM/r1vIU0HBvw1XoJrc7ecN/JYrxYtjXtLC7RBYemKYl1U8CB2NXb75KIL79QMBI8Bkg3rSLBZWp48ZI1G21HkHCnSyBtH/sYZnmASMxfwD2YQTXntC6snZz0PVnQNzdOPo8y1q/Xyp+PhstrrQ5ew= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com; spf=pass smtp.mailfrom=marvell.com; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b=krn6zqqT; arc=none smtp.client-ip=67.231.156.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=marvell.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b="krn6zqqT" Received: from pps.filterd (m0045851.ppops.net [127.0.0.1]) by mx0b-0016f401.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 484HabNu000383; Wed, 4 Sep 2024 11:26:34 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pfpt0220; bh=D ciDPeALtUxlyouw/DQtMs9+ZudZyBWNEmXaMQsmprU=; b=krn6zqqTOWYtMMpCD mLqZKVhTQjdKVji3CIHcZAS+LLYcwSCPch08Ch52BPeB4JYtEV+pSDQUPDGDr5Tr 4CdgMqGWm9w+oBR+4CDGJ3g5tGI9JPDXcOltCmq2bQyKjVlGj6c815TymhfpFA2W fuNGhoR/yqNoGYI1nFQK8hKzcs7BStgSWomJwuFpGlcse1K5ZJNy+klDhwZbUSXG 8RU8G+R193vweQuQ70FnwddOLeo7sHFdub3+NKA6c3rNyJ1rMr+HV8MiV5BvmJBu rUA+wV/aMRQTkU+5f4W6aRpVUM8tivY56iE0SUp+6GL5W316HJFq1+MNHDMs/Fyo fySIA== Received: from dc5-exch05.marvell.com ([199.233.59.128]) by mx0b-0016f401.pphosted.com (PPS) with ESMTPS id 41ev31r6fq-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 04 Sep 2024 11:26:34 -0700 (PDT) Received: from DC5-EXCH05.marvell.com (10.69.176.209) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.4; Wed, 4 Sep 2024 11:26:32 -0700 Received: from maili.marvell.com (10.69.176.80) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server id 15.2.1544.4 via Frontend Transport; Wed, 4 Sep 2024 11:26:32 -0700 Received: from IPBU-BLR-SERVER1.marvell.com (IPBU-BLR-SERVER1.marvell.com [10.28.8.41]) by maili.marvell.com (Postfix) with ESMTP id 103713F7044; Wed, 4 Sep 2024 11:26:29 -0700 (PDT) From: Gowthami Thiagarajan To: , , , CC: , , , Gowthami Thiagarajan Subject: [PATCH v7 2/6] perf/marvell: Refactor to extract platform specific ops - no functional change Date: Wed, 4 Sep 2024 23:56:01 +0530 Message-ID: <20240904182605.953927-3-gthiagarajan@marvell.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240904182605.953927-1-gthiagarajan@marvell.com> References: <20240904182605.953927-1-gthiagarajan@marvell.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Proofpoint-ORIG-GUID: R2BWzE2-2tJ72FGliC7jnN-8LUUOLtXA X-Proofpoint-GUID: R2BWzE2-2tJ72FGliC7jnN-8LUUOLtXA X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-04_16,2024-09-04_01,2024-09-02_01 Content-Type: text/plain; charset="utf-8" This commit introduces a refactor to the Marvell DDR pmu driver, specifically targeting the extraction of ops (referred to as "pmu ops") from the existing driver. Signed-off-by: Gowthami Thiagarajan --- drivers/perf/marvell_cn10k_ddr_pmu.c | 106 +++++++++++++++++++++------ 1 file changed, 83 insertions(+), 23 deletions(-) diff --git a/drivers/perf/marvell_cn10k_ddr_pmu.c b/drivers/perf/marvell_cn= 10k_ddr_pmu.c index e33d383aa6d2..648ad3a740bf 100644 --- a/drivers/perf/marvell_cn10k_ddr_pmu.c +++ b/drivers/perf/marvell_cn10k_ddr_pmu.c @@ -136,6 +136,16 @@ struct cn10k_ddr_pmu { struct hlist_node node; }; =20 +struct ddr_pmu_ops { + void (*enable_read_freerun_counter)(struct cn10k_ddr_pmu *pmu, + bool enable); + void (*enable_write_freerun_counter)(struct cn10k_ddr_pmu *pmu, + bool enable); + void (*clear_read_freerun_counter)(struct cn10k_ddr_pmu *pmu); + void (*clear_write_freerun_counter)(struct cn10k_ddr_pmu *pmu); + void (*pmu_overflow_handler)(struct cn10k_ddr_pmu *pmu, int evt_idx); +}; + #define to_cn10k_ddr_pmu(p) container_of(p, struct cn10k_ddr_pmu, pmu) =20 struct ddr_pmu_platform_data { @@ -152,6 +162,7 @@ struct ddr_pmu_platform_data { u64 ddrc_perf_cnt_freerun_clr; u64 ddrc_perf_cnt_value_wr_op; u64 ddrc_perf_cnt_value_rd_op; + const struct ddr_pmu_ops *ops; }; =20 static ssize_t cn10k_ddr_pmu_event_show(struct device *dev, @@ -375,6 +386,7 @@ static void cn10k_ddr_perf_counter_enable(struct cn10k_= ddr_pmu *pmu, int counter, bool enable) { const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + const struct ddr_pmu_ops *ops =3D p_data->ops; u32 reg; u64 val; =20 @@ -394,21 +406,10 @@ static void cn10k_ddr_perf_counter_enable(struct cn10= k_ddr_pmu *pmu, =20 writeq_relaxed(val, pmu->base + reg); } else { - val =3D readq_relaxed(pmu->base + - p_data->ddrc_perf_cnt_freerun_en); - if (enable) { - if (counter =3D=3D DDRC_PERF_READ_COUNTER_IDX) - val |=3D DDRC_PERF_FREERUN_READ_EN; - else - val |=3D DDRC_PERF_FREERUN_WRITE_EN; - } else { - if (counter =3D=3D DDRC_PERF_READ_COUNTER_IDX) - val &=3D ~DDRC_PERF_FREERUN_READ_EN; - else - val &=3D ~DDRC_PERF_FREERUN_WRITE_EN; - } - writeq_relaxed(val, pmu->base + - p_data->ddrc_perf_cnt_freerun_en); + if (counter =3D=3D DDRC_PERF_READ_COUNTER_IDX) + ops->enable_read_freerun_counter(pmu, enable); + else + ops->enable_write_freerun_counter(pmu, enable); } } =20 @@ -464,6 +465,7 @@ static int cn10k_ddr_perf_event_add(struct perf_event *= event, int flags) { struct cn10k_ddr_pmu *pmu =3D to_cn10k_ddr_pmu(event->pmu); const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + const struct ddr_pmu_ops *ops =3D p_data->ops; struct hw_perf_event *hwc =3D &event->hw; u8 config =3D event->attr.config; int counter, ret; @@ -492,12 +494,9 @@ static int cn10k_ddr_perf_event_add(struct perf_event = *event, int flags) } else { /* fixed event counter, clear counter value */ if (counter =3D=3D DDRC_PERF_READ_COUNTER_IDX) - val =3D DDRC_FREERUN_READ_CNT_CLR; + ops->clear_read_freerun_counter(pmu); else - val =3D DDRC_FREERUN_WRITE_CNT_CLR; - - writeq_relaxed(val, pmu->base + - p_data->ddrc_perf_cnt_freerun_ctrl); + ops->clear_write_freerun_counter(pmu); } =20 hwc->state |=3D PERF_HES_STOPPED; @@ -579,9 +578,63 @@ static void cn10k_ddr_perf_event_update_all(struct cn1= 0k_ddr_pmu *pmu) } } =20 +static void ddr_pmu_enable_read_freerun(struct cn10k_ddr_pmu *pmu, bool en= able) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D readq_relaxed(pmu->base + p_data->ddrc_perf_cnt_freerun_en); + if (enable) + val |=3D DDRC_PERF_FREERUN_READ_EN; + else + val &=3D ~DDRC_PERF_FREERUN_READ_EN; + + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_en); +} + +static void ddr_pmu_enable_write_freerun(struct cn10k_ddr_pmu *pmu, bool e= nable) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D readq_relaxed(pmu->base + p_data->ddrc_perf_cnt_freerun_en); + if (enable) + val |=3D DDRC_PERF_FREERUN_WRITE_EN; + else + val &=3D ~DDRC_PERF_FREERUN_WRITE_EN; + + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_en); +} + +static void ddr_pmu_read_clear_freerun(struct cn10k_ddr_pmu *pmu) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D DDRC_FREERUN_READ_CNT_CLR; + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_ctrl); +} + +static void ddr_pmu_write_clear_freerun(struct cn10k_ddr_pmu *pmu) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D DDRC_FREERUN_WRITE_CNT_CLR; + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_ctrl); +} + +static void ddr_pmu_overflow_hander(struct cn10k_ddr_pmu *pmu, int evt_idx) +{ + cn10k_ddr_perf_event_update_all(pmu); + cn10k_ddr_perf_pmu_disable(&pmu->pmu); + cn10k_ddr_perf_pmu_enable(&pmu->pmu); +} + static irqreturn_t cn10k_ddr_pmu_overflow_handler(struct cn10k_ddr_pmu *pm= u) { const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + const struct ddr_pmu_ops *ops =3D p_data->ops; struct perf_event *event; struct hw_perf_event *hwc; u64 prev_count, new_count; @@ -621,9 +674,7 @@ static irqreturn_t cn10k_ddr_pmu_overflow_handler(struc= t cn10k_ddr_pmu *pmu) value =3D cn10k_ddr_perf_read_counter(pmu, i); if (value =3D=3D p_data->counter_max_val) { pr_info("Counter-(%d) reached max value\n", i); - cn10k_ddr_perf_event_update_all(pmu); - cn10k_ddr_perf_pmu_disable(&pmu->pmu); - cn10k_ddr_perf_pmu_enable(&pmu->pmu); + ops->pmu_overflow_handler(pmu, i); } } =20 @@ -662,6 +713,14 @@ static int cn10k_ddr_pmu_offline_cpu(unsigned int cpu,= struct hlist_node *node) return 0; } =20 +static const struct ddr_pmu_ops ddr_pmu_ops =3D { + .enable_read_freerun_counter =3D ddr_pmu_enable_read_freerun, + .enable_write_freerun_counter =3D ddr_pmu_enable_write_freerun, + .clear_read_freerun_counter =3D ddr_pmu_read_clear_freerun, + .clear_write_freerun_counter =3D ddr_pmu_write_clear_freerun, + .pmu_overflow_handler =3D ddr_pmu_overflow_hander, +}; + static const struct ddr_pmu_platform_data cn10k_ddr_pmu_pdata =3D { .counter_overflow_val =3D BIT_ULL(48), .counter_max_val =3D GENMASK_ULL(48, 0), @@ -676,6 +735,7 @@ static const struct ddr_pmu_platform_data cn10k_ddr_pmu= _pdata =3D { .ddrc_perf_cnt_freerun_clr =3D 0, .ddrc_perf_cnt_value_wr_op =3D CN10K_DDRC_PERF_CNT_VALUE_WR_OP, .ddrc_perf_cnt_value_rd_op =3D CN10K_DDRC_PERF_CNT_VALUE_RD_OP, + .ops =3D &ddr_pmu_ops, }; =20 static int cn10k_ddr_perf_probe(struct platform_device *pdev) --=20 2.25.1 From nobody Fri Dec 19 16:53:31 2025 Received: from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com [67.231.156.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id EEC151E412E for ; Wed, 4 Sep 2024 18:26:45 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=67.231.156.173 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474407; cv=none; b=XiNBpE+OMoquwpWYfb7e0TazTLeZAWO9Q+4GGJA2cpRDWo5jAAmM2PuMOBZvM1HRC36zPA9O+kI4HU8GIMqPX2IGJA/nb4ocHpaVxW8pmNXWgBl/+BfHvJKFKQZ7IAayaWoSF/2Gm/qEBmYy32kGN5DQZJXGQVBUMdQ1DsxxOMM= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474407; c=relaxed/simple; bh=Nsc/Ov81UHgdnQynWWoSifpTaJXPqz7CqaQsqehchk0=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=Pzw9rvkn78+M68IvJ/mjBRwIJ3VEvBSimQvrCrciXJkEDXSaLcKJUgwaKlZtrapNtjokb5j1ZOwq4xHNylz/F4Fa1png4D3suK5OrTg2c2ym+NshTfYMTUywtXkQaFjFSUVopMbvZFmj+MLtchxkY1dTMzgkndSD0ValHk0zkaU= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com; spf=pass smtp.mailfrom=marvell.com; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b=Xq6jvANo; arc=none smtp.client-ip=67.231.156.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=marvell.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b="Xq6jvANo" Received: from pps.filterd (m0045851.ppops.net [127.0.0.1]) by mx0b-0016f401.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 484HZsQa031904; Wed, 4 Sep 2024 11:26:38 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pfpt0220; bh=8 /kL6cuJizHe7Bia8eHKEHtVx3MtTcCmYU+ozWWrc2o=; b=Xq6jvANo8jmSqgHZ/ YzfqJZiuEnl1ivjd0dyJw31v9uUxgAgtRX4NNRHGvlsWWGa2e7FtPXsfziVMGf1J z5UlUJg0tCWGcjAtW5+Luqb4TFhVSstUor+CUMjpS0ONauazof+RdS7ZMbfFiOVs DSDb77K9v8GbggUOzRWaw8WTSXDjjMlw/Fu73A48XDOWqfXmAfztoQUPJyuqVMot zRAEU+Tn1WsyNleOmYVQdwseHFFZ6vNavYHEE7RIZoGhBC/K5e3aIHoR2uZsse1H LhsCkpeej5Ms9c5n4BKj4Md8AOXpbdfHXAZf/rXadPIA1rdRBFl8qJILgAzX88QY a1R7A== Received: from dc5-exch05.marvell.com ([199.233.59.128]) by mx0b-0016f401.pphosted.com (PPS) with ESMTPS id 41ev31r6fx-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 04 Sep 2024 11:26:38 -0700 (PDT) Received: from DC5-EXCH05.marvell.com (10.69.176.209) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.4; Wed, 4 Sep 2024 11:26:36 -0700 Received: from maili.marvell.com (10.69.176.80) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server id 15.2.1544.4 via Frontend Transport; Wed, 4 Sep 2024 11:26:36 -0700 Received: from IPBU-BLR-SERVER1.marvell.com (IPBU-BLR-SERVER1.marvell.com [10.28.8.41]) by maili.marvell.com (Postfix) with ESMTP id 244D43F7044; Wed, 4 Sep 2024 11:26:33 -0700 (PDT) From: Gowthami Thiagarajan To: , , , CC: , , , Gowthami Thiagarajan Subject: [PATCH v7 3/6] perf/marvell: Refactor to add version - no functional change Date: Wed, 4 Sep 2024 23:56:02 +0530 Message-ID: <20240904182605.953927-4-gthiagarajan@marvell.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240904182605.953927-1-gthiagarajan@marvell.com> References: <20240904182605.953927-1-gthiagarajan@marvell.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Proofpoint-ORIG-GUID: VaklIAF0pxENqn4E0KwJ-8gzGZuh9zsM X-Proofpoint-GUID: VaklIAF0pxENqn4E0KwJ-8gzGZuh9zsM X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-04_16,2024-09-04_01,2024-09-02_01 Content-Type: text/plain; charset="utf-8" This change is aimed at improving the maintainability of the code and laying the groundwork for versioning within the driver. No functional changes are introduced in this commit; the driver's behavior and performance remain unchanged. Signed-off-by: Gowthami Thiagarajan --- drivers/perf/marvell_cn10k_ddr_pmu.c | 61 ++++++++++++++++++---------- 1 file changed, 40 insertions(+), 21 deletions(-) diff --git a/drivers/perf/marvell_cn10k_ddr_pmu.c b/drivers/perf/marvell_cn= 10k_ddr_pmu.c index 648ad3a740bf..86e63a2b02d7 100644 --- a/drivers/perf/marvell_cn10k_ddr_pmu.c +++ b/drivers/perf/marvell_cn10k_ddr_pmu.c @@ -124,10 +124,19 @@ #define CN10K_DDRC_PERF_CNT_VALUE_WR_OP 0x80D0 #define CN10K_DDRC_PERF_CNT_VALUE_RD_OP 0x80D8 =20 +enum mrvl_ddr_pmu_version { + DDR_PMU_V1 =3D 1, +}; + +struct ddr_pmu_data { + int id; +}; + struct cn10k_ddr_pmu { struct pmu pmu; void __iomem *base; const struct ddr_pmu_platform_data *p_data; + int version; unsigned int cpu; struct device *dev; int active_events; @@ -738,12 +747,17 @@ static const struct ddr_pmu_platform_data cn10k_ddr_p= mu_pdata =3D { .ops =3D &ddr_pmu_ops, }; =20 +static const struct ddr_pmu_data ddr_pmu_data =3D { + .id =3D DDR_PMU_V1, +}; + static int cn10k_ddr_perf_probe(struct platform_device *pdev) { const struct ddr_pmu_data *dev_data; struct cn10k_ddr_pmu *ddr_pmu; struct resource *res; void __iomem *base; + int version; char *name; int ret; =20 @@ -760,31 +774,36 @@ static int cn10k_ddr_perf_probe(struct platform_devic= e *pdev) return -ENODEV; } =20 + version =3D dev_data->id; + ddr_pmu->version =3D version; + base =3D devm_platform_get_and_ioremap_resource(pdev, 0, &res); if (IS_ERR(base)) return PTR_ERR(base); =20 ddr_pmu->base =3D base; =20 - ddr_pmu->p_data =3D &cn10k_ddr_pmu_pdata; - /* Setup the PMU counter to work in manual mode */ - writeq_relaxed(OP_MODE_CTRL_VAL_MANUAL, ddr_pmu->base + - ddr_pmu->p_data->ddrc_perf_cnt_op_mode_ctrl); - - ddr_pmu->pmu =3D (struct pmu) { - .module =3D THIS_MODULE, - .capabilities =3D PERF_PMU_CAP_NO_EXCLUDE, - .task_ctx_nr =3D perf_invalid_context, - .attr_groups =3D cn10k_attr_groups, - .event_init =3D cn10k_ddr_perf_event_init, - .add =3D cn10k_ddr_perf_event_add, - .del =3D cn10k_ddr_perf_event_del, - .start =3D cn10k_ddr_perf_event_start, - .stop =3D cn10k_ddr_perf_event_stop, - .read =3D cn10k_ddr_perf_event_update, - .pmu_enable =3D cn10k_ddr_perf_pmu_enable, - .pmu_disable =3D cn10k_ddr_perf_pmu_disable, - }; + if (version =3D=3D DDR_PMU_V1) { + ddr_pmu->p_data =3D &cn10k_ddr_pmu_pdata; + /* Setup the PMU counter to work in manual mode */ + writeq_relaxed(OP_MODE_CTRL_VAL_MANUAL, ddr_pmu->base + + ddr_pmu->p_data->ddrc_perf_cnt_op_mode_ctrl); + + ddr_pmu->pmu =3D (struct pmu) { + .module =3D THIS_MODULE, + .capabilities =3D PERF_PMU_CAP_NO_EXCLUDE, + .task_ctx_nr =3D perf_invalid_context, + .attr_groups =3D cn10k_attr_groups, + .event_init =3D cn10k_ddr_perf_event_init, + .add =3D cn10k_ddr_perf_event_add, + .del =3D cn10k_ddr_perf_event_del, + .start =3D cn10k_ddr_perf_event_start, + .stop =3D cn10k_ddr_perf_event_stop, + .read =3D cn10k_ddr_perf_event_update, + .pmu_enable =3D cn10k_ddr_perf_pmu_enable, + .pmu_disable =3D cn10k_ddr_perf_pmu_disable, + }; + } =20 /* Choose this cpu to collect perf data */ ddr_pmu->cpu =3D raw_smp_processor_id(); @@ -827,7 +846,7 @@ static void cn10k_ddr_perf_remove(struct platform_devic= e *pdev) =20 #ifdef CONFIG_OF static const struct of_device_id cn10k_ddr_pmu_of_match[] =3D { - { .compatible =3D "marvell,cn10k-ddr-pmu", }, + { .compatible =3D "marvell,cn10k-ddr-pmu", .data =3D &ddr_pmu_data}, { }, }; MODULE_DEVICE_TABLE(of, cn10k_ddr_pmu_of_match); @@ -835,7 +854,7 @@ MODULE_DEVICE_TABLE(of, cn10k_ddr_pmu_of_match); =20 #ifdef CONFIG_ACPI static const struct acpi_device_id cn10k_ddr_pmu_acpi_match[] =3D { - {"MRVL000A", 0}, + {"MRVL000A", (kernel_ulong_t)&ddr_pmu_data}, {}, }; MODULE_DEVICE_TABLE(acpi, cn10k_ddr_pmu_acpi_match); --=20 2.25.1 From nobody Fri Dec 19 16:53:31 2025 Received: from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com [67.231.156.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 7283B1E2025 for ; Wed, 4 Sep 2024 18:26:52 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=67.231.156.173 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474414; cv=none; b=lzfZdTENUy9HrYDIWlgY7ePDbXX+Zpf7Z/eAtA6rNTH8vDGtFxbMVPkwJu4NLt2n/JtRIBzCYWpbVwcucKJprZp0y0o/40VNhrC9ghF4hB17E7hXFnnDVoPJOqX4Dgl8yOWtuDQGBz7D9/L3rSbgp1Ol+fJYXSW4YGxjP3lTRbE= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474414; c=relaxed/simple; bh=kqq8pT1LP3S6vactUZKET6rzd/7b2EFe9Eaj6lWU3G8=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=FDPkF4DK1uRBxMCQ5i6Nyo9WIyW3/2In5/T0Pk74ZZmb716oQUcyXO35pZyoQ86KHCK/8ctQ935YxXMyMaxUd5aaGC0bDwJNeuzzm7UpAIfcYBzmaPRStIQGNbHbpiSS9R0dsr6jahgRnZkgYAVCbaMUfWZN4lRvZPNncHcXO/g= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com; spf=pass smtp.mailfrom=marvell.com; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b=d2kPm73O; arc=none smtp.client-ip=67.231.156.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=marvell.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b="d2kPm73O" Received: from pps.filterd (m0045851.ppops.net [127.0.0.1]) by mx0b-0016f401.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 484HZw48031921; Wed, 4 Sep 2024 11:26:42 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pfpt0220; bh=R u/70uB/NxGKFHpUpJGV5ZaNpkmJbpOla3Md1PvTZdM=; b=d2kPm73OGHBCyGwbo WZ8h4lDeCN5aCXCSMDudoEhCnKjI/ahEzbAij96qfF1CT/ctUILG1YMm7jTR/rol DYlasHVrFOj3Luclon+lvyqOAiBUbulrd3DiPhHeFN74Xg26XXOg+Ci0j2vPGjag m7GmZYSWEotkX+N2iNI7PotFdqFf9PkdP8Ffch+IJ9DjvFmnp8FC7BcdxJzRfW1p qnHHnrq2Y09yvz2ZPHdzdU88uOJGrZZxur4Bft8GOocArqhWARNskuenMG4Hn1BG gOEEMOvxGafdZmDnaVE4Xy0ycLcFFbTej7d4RAsI2HJ8uVgqB9191SR5qVqn3tKo Acj6g== Received: from dc5-exch05.marvell.com ([199.233.59.128]) by mx0b-0016f401.pphosted.com (PPS) with ESMTPS id 41ev31r6g3-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 04 Sep 2024 11:26:42 -0700 (PDT) Received: from DC5-EXCH05.marvell.com (10.69.176.209) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.4; Wed, 4 Sep 2024 11:26:40 -0700 Received: from maili.marvell.com (10.69.176.80) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server id 15.2.1544.4 via Frontend Transport; Wed, 4 Sep 2024 11:26:40 -0700 Received: from IPBU-BLR-SERVER1.marvell.com (IPBU-BLR-SERVER1.marvell.com [10.28.8.41]) by maili.marvell.com (Postfix) with ESMTP id 3B3FF3F7044; Wed, 4 Sep 2024 11:26:37 -0700 (PDT) From: Gowthami Thiagarajan To: , , , CC: , , , Gowthami Thiagarajan Subject: [PATCH v7 4/6] perf/marvell: Odyssey DDR Performance monitor support Date: Wed, 4 Sep 2024 23:56:03 +0530 Message-ID: <20240904182605.953927-5-gthiagarajan@marvell.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240904182605.953927-1-gthiagarajan@marvell.com> References: <20240904182605.953927-1-gthiagarajan@marvell.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Proofpoint-ORIG-GUID: PVWyhfnAQqs1xpAjfu6JLIc5FCJhIiUe X-Proofpoint-GUID: PVWyhfnAQqs1xpAjfu6JLIc5FCJhIiUe X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-04_16,2024-09-04_01,2024-09-02_01 Content-Type: text/plain; charset="utf-8" Odyssey DRAM Subsystem supports eight counters for monitoring performance and software can program those counters to monitor any of the defined performance events. Supported performance events include those counted at the interface between the DDR controller and the PHY, interface between the DDR Controller and the CHI interconnect, or within the DDR Controller. Additionally DSS also supports two fixed performance event counters, one for ddr reads and the other for ddr writes. Signed-off-by: Gowthami Thiagarajan --- Documentation/admin-guide/perf/index.rst | 1 + .../admin-guide/perf/mrvl-odyssey-ddr-pmu.rst | 80 ++++++ drivers/perf/marvell_cn10k_ddr_pmu.c | 257 +++++++++++++++++- 3 files changed, 335 insertions(+), 3 deletions(-) create mode 100644 Documentation/admin-guide/perf/mrvl-odyssey-ddr-pmu.rst diff --git a/Documentation/admin-guide/perf/index.rst b/Documentation/admin= -guide/perf/index.rst index 7eb3dcd6f4da..d673ccfea903 100644 --- a/Documentation/admin-guide/perf/index.rst +++ b/Documentation/admin-guide/perf/index.rst @@ -14,6 +14,7 @@ Performance monitor support qcom_l2_pmu qcom_l3_pmu starfive_starlink_pmu + mrvl-odyssey-ddr-pmu arm-ccn arm-cmn xgene-pmu diff --git a/Documentation/admin-guide/perf/mrvl-odyssey-ddr-pmu.rst b/Docu= mentation/admin-guide/perf/mrvl-odyssey-ddr-pmu.rst new file mode 100644 index 000000000000..2e817593a4d9 --- /dev/null +++ b/Documentation/admin-guide/perf/mrvl-odyssey-ddr-pmu.rst @@ -0,0 +1,80 @@ +=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D +Marvell Odyssey DDR PMU Performance Monitoring Unit (PMU UNCORE) +=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D + +Odyssey DRAM Subsystem supports eight counters for monitoring performance +and software can program those counters to monitor any of the defined +performance events. Supported performance events include those counted +at the interface between the DDR controller and the PHY, interface between +the DDR Controller and the CHI interconnect, or within the DDR Controller. + +Additionally DSS also supports two fixed performance event counters, one +for ddr reads and the other for ddr writes. + +The counter will be operating in either manual or auto mode. + +The PMU driver exposes the available events and format options under sysfs= :: + + /sys/bus/event_source/devices/mrvl_ddr_pmu_<>/events/ + /sys/bus/event_source/devices/mrvl_ddr_pmu_<>/format/ + +Examples:: + + $ perf list | grep ddr + mrvl_ddr_pmu_<>/ddr_act_bypass_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_bsm_alloc/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_bsm_starvation/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_cam_active_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_cam_mwr/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_cam_rd_active_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_cam_rd_or_wr_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_cam_read/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_cam_wr_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_cam_write/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_capar_error/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_crit_ref/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_ddr_reads/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_ddr_writes/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_dfi_cmd_is_retry/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_dfi_cycles/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_dfi_parity_poison/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_dfi_rd_data_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_dfi_wr_data_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_dqsosc_mpc/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_dqsosc_mrr/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_enter_mpsm/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_enter_powerdown/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_enter_selfref/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_hif_pri_rdaccess/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_hif_rd_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_hif_rd_or_wr_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_hif_rmw_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_hif_wr_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_hpri_sched_rd_crit_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_load_mode/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_lpri_sched_rd_crit_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_precharge/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_precharge_for_other/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_precharge_for_rdwr/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_raw_hazard/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_rd_bypass_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_rd_crc_error/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_rd_uc_ecc_error/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_rdwr_transitions/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_refresh/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_retry_fifo_full/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_spec_ref/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_tcr_mrr/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_war_hazard/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_waw_hazard/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_win_limit_reached_rd/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_win_limit_reached_wr/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_wr_crc_error/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_wr_trxn_crit_access/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_write_combine/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_zqcl/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_zqlatch/ [Kernel PMU event] + mrvl_ddr_pmu_<>/ddr_zqstart/ [Kernel PMU event] + + $ perf stat -e ddr_cam_read,ddr_cam_write,ddr_cam_active_access,dd= r_cam + rd_or_wr_access,ddr_cam_rd_active_access,ddr_cam_mwr diff --git a/drivers/perf/marvell_cn10k_ddr_pmu.c b/drivers/perf/marvell_cn= 10k_ddr_pmu.c index 86e63a2b02d7..533e84f33a8f 100644 --- a/drivers/perf/marvell_cn10k_ddr_pmu.c +++ b/drivers/perf/marvell_cn10k_ddr_pmu.c @@ -16,23 +16,28 @@ =20 /* Performance Counters Operating Mode Control Registers */ #define CN10K_DDRC_PERF_CNT_OP_MODE_CTRL 0x8020 +#define ODY_DDRC_PERF_CNT_OP_MODE_CTRL 0x20020 #define OP_MODE_CTRL_VAL_MANUAL 0x1 =20 /* Performance Counters Start Operation Control Registers */ #define CN10K_DDRC_PERF_CNT_START_OP_CTRL 0x8028 +#define ODY_DDRC_PERF_CNT_START_OP_CTRL 0x200A0 #define START_OP_CTRL_VAL_START 0x1ULL #define START_OP_CTRL_VAL_ACTIVE 0x2 =20 /* Performance Counters End Operation Control Registers */ #define CN10K_DDRC_PERF_CNT_END_OP_CTRL 0x8030 +#define ODY_DDRC_PERF_CNT_END_OP_CTRL 0x200E0 #define END_OP_CTRL_VAL_END 0x1ULL =20 /* Performance Counters End Status Registers */ #define CN10K_DDRC_PERF_CNT_END_STATUS 0x8038 +#define ODY_DDRC_PERF_CNT_END_STATUS 0x20120 #define END_STATUS_VAL_END_TIMER_MODE_END 0x1 =20 /* Performance Counters Configuration Registers */ #define CN10K_DDRC_PERF_CFG_BASE 0x8040 +#define ODY_DDRC_PERF_CFG_BASE 0x20160 =20 /* 8 Generic event counter + 2 fixed event counters */ #define DDRC_PERF_NUM_GEN_COUNTERS 8 @@ -57,6 +62,13 @@ * program event bitmap in h/w. * */ +#define EVENT_DFI_CMD_IS_RETRY 61 +#define EVENT_RD_UC_ECC_ERROR 60 +#define EVENT_RD_CRC_ERROR 59 +#define EVENT_CAPAR_ERROR 58 +#define EVENT_WR_CRC_ERROR 57 +#define EVENT_DFI_PARITY_POISON 56 + #define EVENT_OP_IS_ZQLATCH 55 #define EVENT_OP_IS_ZQSTART 54 #define EVENT_OP_IS_TCR_MRR 53 @@ -106,6 +118,7 @@ =20 /* Event counter value registers */ #define CN10K_DDRC_PERF_CNT_VALUE_BASE 0x8080 +#define ODY_DDRC_PERF_CNT_VALUE_BASE 0x201C0 =20 /* Fixed event counter enable/disable register */ #define CN10K_DDRC_PERF_CNT_FREERUN_EN 0x80C0 @@ -114,18 +127,25 @@ =20 /* Fixed event counter control register */ #define CN10K_DDRC_PERF_CNT_FREERUN_CTRL 0x80C8 +#define ODY_DDRC_PERF_CNT_FREERUN_CTRL 0x20240 #define DDRC_FREERUN_WRITE_CNT_CLR 0x1 #define DDRC_FREERUN_READ_CNT_CLR 0x2 =20 +/* Fixed event counter clear register, defined only for Odyssey */ +#define ODY_DDRC_PERF_CNT_FREERUN_CLR 0x20248 + #define DDRC_PERF_CNT_VALUE_OVERFLOW BIT_ULL(48) #define DDRC_PERF_CNT_MAX_VALUE GENMASK_ULL(48, 0) =20 /* Fixed event counter value register */ #define CN10K_DDRC_PERF_CNT_VALUE_WR_OP 0x80D0 #define CN10K_DDRC_PERF_CNT_VALUE_RD_OP 0x80D8 +#define ODY_DDRC_PERF_CNT_VALUE_WR_OP 0x20250 +#define ODY_DDRC_PERF_CNT_VALUE_RD_OP 0x20258 =20 enum mrvl_ddr_pmu_version { DDR_PMU_V1 =3D 1, + DDR_PMU_V2, }; =20 struct ddr_pmu_data { @@ -249,6 +269,85 @@ static struct attribute *cn10k_ddr_perf_events_attrs[]= =3D { NULL }; =20 +static struct attribute *odyssey_ddr_perf_events_attrs[] =3D { + /* Programmable */ + CN10K_DDR_PMU_EVENT_ATTR(ddr_hif_rd_or_wr_access, EVENT_HIF_RD_OR_WR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_hif_wr_access, EVENT_HIF_WR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_hif_rd_access, EVENT_HIF_RD), + CN10K_DDR_PMU_EVENT_ATTR(ddr_hif_rmw_access, EVENT_HIF_RMW), + CN10K_DDR_PMU_EVENT_ATTR(ddr_hif_pri_rdaccess, EVENT_HIF_HI_PRI_RD), + CN10K_DDR_PMU_EVENT_ATTR(ddr_rd_bypass_access, EVENT_READ_BYPASS), + CN10K_DDR_PMU_EVENT_ATTR(ddr_act_bypass_access, EVENT_ACT_BYPASS), + CN10K_DDR_PMU_EVENT_ATTR(ddr_dfi_wr_data_access, + EVENT_DFI_WR_DATA_CYCLES), + CN10K_DDR_PMU_EVENT_ATTR(ddr_dfi_rd_data_access, + EVENT_DFI_RD_DATA_CYCLES), + CN10K_DDR_PMU_EVENT_ATTR(ddr_hpri_sched_rd_crit_access, + EVENT_HPR_XACT_WHEN_CRITICAL), + CN10K_DDR_PMU_EVENT_ATTR(ddr_lpri_sched_rd_crit_access, + EVENT_LPR_XACT_WHEN_CRITICAL), + CN10K_DDR_PMU_EVENT_ATTR(ddr_wr_trxn_crit_access, + EVENT_WR_XACT_WHEN_CRITICAL), + CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_active_access, EVENT_OP_IS_ACTIVATE), + CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_rd_or_wr_access, + EVENT_OP_IS_RD_OR_WR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_rd_active_access, + EVENT_OP_IS_RD_ACTIVATE), + CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_read, EVENT_OP_IS_RD), + CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_write, EVENT_OP_IS_WR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_mwr, EVENT_OP_IS_MWR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_precharge, EVENT_OP_IS_PRECHARGE), + CN10K_DDR_PMU_EVENT_ATTR(ddr_precharge_for_rdwr, + EVENT_PRECHARGE_FOR_RDWR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_precharge_for_other, + EVENT_PRECHARGE_FOR_OTHER), + CN10K_DDR_PMU_EVENT_ATTR(ddr_rdwr_transitions, EVENT_RDWR_TRANSITIONS), + CN10K_DDR_PMU_EVENT_ATTR(ddr_write_combine, EVENT_WRITE_COMBINE), + CN10K_DDR_PMU_EVENT_ATTR(ddr_war_hazard, EVENT_WAR_HAZARD), + CN10K_DDR_PMU_EVENT_ATTR(ddr_raw_hazard, EVENT_RAW_HAZARD), + CN10K_DDR_PMU_EVENT_ATTR(ddr_waw_hazard, EVENT_WAW_HAZARD), + CN10K_DDR_PMU_EVENT_ATTR(ddr_enter_selfref, EVENT_OP_IS_ENTER_SELFREF), + CN10K_DDR_PMU_EVENT_ATTR(ddr_enter_powerdown, + EVENT_OP_IS_ENTER_POWERDOWN), + CN10K_DDR_PMU_EVENT_ATTR(ddr_enter_mpsm, EVENT_OP_IS_ENTER_MPSM), + CN10K_DDR_PMU_EVENT_ATTR(ddr_refresh, EVENT_OP_IS_REFRESH), + CN10K_DDR_PMU_EVENT_ATTR(ddr_crit_ref, EVENT_OP_IS_CRIT_REF), + CN10K_DDR_PMU_EVENT_ATTR(ddr_spec_ref, EVENT_OP_IS_SPEC_REF), + CN10K_DDR_PMU_EVENT_ATTR(ddr_load_mode, EVENT_OP_IS_LOAD_MODE), + CN10K_DDR_PMU_EVENT_ATTR(ddr_zqcl, EVENT_OP_IS_ZQCL), + CN10K_DDR_PMU_EVENT_ATTR(ddr_cam_wr_access, EVENT_OP_IS_ZQCS), + CN10K_DDR_PMU_EVENT_ATTR(ddr_dfi_cycles, EVENT_DFI_OR_HPR_REQ_NOCRED), + CN10K_DDR_PMU_EVENT_ATTR(ddr_retry_fifo_full, + EVENT_RETRY_FIFO_FULL_OR_LPR_REQ_NOCRED), + CN10K_DDR_PMU_EVENT_ATTR(ddr_bsm_alloc, EVENT_BSM_ALLOC), + CN10K_DDR_PMU_EVENT_ATTR(ddr_bsm_starvation, EVENT_BSM_STARVATION), + CN10K_DDR_PMU_EVENT_ATTR(ddr_win_limit_reached_rd, + EVENT_VISIBLE_WIN_LIMIT_REACHED_RD), + CN10K_DDR_PMU_EVENT_ATTR(ddr_win_limit_reached_wr, + EVENT_VISIBLE_WIN_LIMIT_REACHED_WR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_dqsosc_mpc, EVENT_OP_IS_DQSOSC_MPC), + CN10K_DDR_PMU_EVENT_ATTR(ddr_dqsosc_mrr, EVENT_OP_IS_DQSOSC_MRR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_tcr_mrr, EVENT_OP_IS_TCR_MRR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_zqstart, EVENT_OP_IS_ZQSTART), + CN10K_DDR_PMU_EVENT_ATTR(ddr_zqlatch, EVENT_OP_IS_ZQLATCH), + CN10K_DDR_PMU_EVENT_ATTR(ddr_dfi_parity_poison, + EVENT_DFI_PARITY_POISON), + CN10K_DDR_PMU_EVENT_ATTR(ddr_wr_crc_error, EVENT_WR_CRC_ERROR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_capar_error, EVENT_CAPAR_ERROR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_rd_crc_error, EVENT_RD_CRC_ERROR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_rd_uc_ecc_error, EVENT_RD_UC_ECC_ERROR), + CN10K_DDR_PMU_EVENT_ATTR(ddr_dfi_cmd_is_retry, EVENT_DFI_CMD_IS_RETRY), + /* Free run event counters */ + CN10K_DDR_PMU_EVENT_ATTR(ddr_ddr_reads, EVENT_DDR_READS), + CN10K_DDR_PMU_EVENT_ATTR(ddr_ddr_writes, EVENT_DDR_WRITES), + NULL +}; + +static struct attribute_group odyssey_ddr_perf_events_attr_group =3D { + .name =3D "events", + .attrs =3D odyssey_ddr_perf_events_attrs, +}; + static struct attribute_group cn10k_ddr_perf_events_attr_group =3D { .name =3D "events", .attrs =3D cn10k_ddr_perf_events_attrs, @@ -294,6 +393,13 @@ static const struct attribute_group *cn10k_attr_groups= [] =3D { NULL, }; =20 +static const struct attribute_group *odyssey_attr_groups[] =3D { + &odyssey_ddr_perf_events_attr_group, + &cn10k_ddr_perf_format_attr_group, + &cn10k_ddr_perf_cpumask_attr_group, + NULL +}; + /* Default poll timeout is 100 sec, which is very sufficient for * 48 bit counter incremented max at 5.6 GT/s, which may take many * hours to overflow. @@ -306,20 +412,27 @@ static ktime_t cn10k_ddr_pmu_timer_period(void) return ms_to_ktime((u64)cn10k_ddr_pmu_poll_period_sec * USEC_PER_SEC); } =20 -static int ddr_perf_get_event_bitmap(int eventid, u64 *event_bitmap) +static int ddr_perf_get_event_bitmap(int eventid, u64 *event_bitmap, + struct cn10k_ddr_pmu *ddr_pmu) { switch (eventid) { case EVENT_HIF_RD_OR_WR ... EVENT_WAW_HAZARD: case EVENT_OP_IS_REFRESH ... EVENT_OP_IS_ZQLATCH: *event_bitmap =3D (1ULL << (eventid - 1)); break; + case EVENT_DFI_PARITY_POISON ...EVENT_DFI_CMD_IS_RETRY: + if (ddr_pmu->version =3D=3D DDR_PMU_V2) + *event_bitmap =3D (1ULL << (eventid - 1)); + else + goto err; + break; case EVENT_OP_IS_ENTER_SELFREF: case EVENT_OP_IS_ENTER_POWERDOWN: case EVENT_OP_IS_ENTER_MPSM: *event_bitmap =3D (0xFULL << (eventid - 1)); break; default: - pr_err("%s Invalid eventid %d\n", __func__, eventid); +err: pr_err("%s Invalid eventid %d\n", __func__, eventid); return -EINVAL; } =20 @@ -457,15 +570,43 @@ static void cn10k_ddr_perf_event_update(struct perf_e= vent *event) local64_add((new_count - prev_count) & mask, &event->count); } =20 +static void cn10k_ddr_perf_counter_start(struct cn10k_ddr_pmu *ddr_pmu, + int counter) +{ + const struct ddr_pmu_platform_data *p_data =3D ddr_pmu->p_data; + u64 ctrl_reg =3D p_data->ddrc_perf_cnt_start_op_ctrl; + + writeq_relaxed(START_OP_CTRL_VAL_START, ddr_pmu->base + + DDRC_PERF_REG(ctrl_reg, counter)); +} + +static void cn10k_ddr_perf_counter_stop(struct cn10k_ddr_pmu *ddr_pmu, + int counter) +{ + const struct ddr_pmu_platform_data *p_data =3D ddr_pmu->p_data; + u64 ctrl_reg =3D p_data->ddrc_perf_cnt_end_op_ctrl; + + writeq_relaxed(END_OP_CTRL_VAL_END, ddr_pmu->base + + DDRC_PERF_REG(ctrl_reg, counter)); +} + static void cn10k_ddr_perf_event_start(struct perf_event *event, int flags) { struct cn10k_ddr_pmu *pmu =3D to_cn10k_ddr_pmu(event->pmu); + u64 ctrl_reg =3D pmu->p_data->ddrc_perf_cnt_op_mode_ctrl; struct hw_perf_event *hwc =3D &event->hw; int counter =3D hwc->idx; =20 local64_set(&hwc->prev_count, 0); =20 cn10k_ddr_perf_counter_enable(pmu, counter, true); + if (pmu->version =3D=3D DDR_PMU_V2) { + /* Setup the PMU counter to work in manual mode */ + writeq_relaxed(OP_MODE_CTRL_VAL_MANUAL, pmu->base + + DDRC_PERF_REG(ctrl_reg, counter)); + + cn10k_ddr_perf_counter_start(pmu, counter); + } =20 hwc->state =3D 0; } @@ -495,7 +636,7 @@ static int cn10k_ddr_perf_event_add(struct perf_event *= event, int flags) if (counter < DDRC_PERF_NUM_GEN_COUNTERS) { /* Generic counters, configure event id */ reg_offset =3D DDRC_PERF_CFG(p_data->ddrc_perf_cfg_base, counter); - ret =3D ddr_perf_get_event_bitmap(config, &val); + ret =3D ddr_perf_get_event_bitmap(config, &val, pmu); if (ret) return ret; =20 @@ -524,6 +665,9 @@ static void cn10k_ddr_perf_event_stop(struct perf_event= *event, int flags) =20 cn10k_ddr_perf_counter_enable(pmu, counter, false); =20 + if (pmu->version =3D=3D DDR_PMU_V2) + cn10k_ddr_perf_counter_stop(pmu, counter); + if (flags & PERF_EF_UPDATE) cn10k_ddr_perf_event_update(event); =20 @@ -640,6 +784,66 @@ static void ddr_pmu_overflow_hander(struct cn10k_ddr_p= mu *pmu, int evt_idx) cn10k_ddr_perf_pmu_enable(&pmu->pmu); } =20 +static void ddr_pmu_v2_enable_read_freerun(struct cn10k_ddr_pmu *pmu, + bool enable) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D readq_relaxed(pmu->base + p_data->ddrc_perf_cnt_freerun_ctrl); + if (enable) + val |=3D DDRC_PERF_FREERUN_READ_EN; + else + val &=3D ~DDRC_PERF_FREERUN_READ_EN; + + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_ctrl); +} + +static void ddr_pmu_v2_enable_write_freerun(struct cn10k_ddr_pmu *pmu, + bool enable) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D readq_relaxed(pmu->base + p_data->ddrc_perf_cnt_freerun_ctrl); + if (enable) + val |=3D DDRC_PERF_FREERUN_WRITE_EN; + else + val &=3D ~DDRC_PERF_FREERUN_WRITE_EN; + + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_ctrl); +} + +static void ddr_pmu_v2_read_clear_freerun(struct cn10k_ddr_pmu *pmu) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D DDRC_FREERUN_READ_CNT_CLR; + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_clr); +} + +static void ddr_pmu_v2_write_clear_freerun(struct cn10k_ddr_pmu *pmu) +{ + const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; + u64 val; + + val =3D DDRC_FREERUN_WRITE_CNT_CLR; + writeq_relaxed(val, pmu->base + p_data->ddrc_perf_cnt_freerun_clr); +} + +static void ddr_pmu_v2_overflow_hander(struct cn10k_ddr_pmu *pmu, int evt_= idx) +{ + /* + * On reaching the maximum value of the counter, the counter freezes + * there. The particular event is updated and the respective counter + * is stopped and started again so that it starts counting from zero + */ + cn10k_ddr_perf_event_update(pmu->events[evt_idx]); + cn10k_ddr_perf_counter_stop(pmu, evt_idx); + cn10k_ddr_perf_counter_start(pmu, evt_idx); +} + static irqreturn_t cn10k_ddr_pmu_overflow_handler(struct cn10k_ddr_pmu *pm= u) { const struct ddr_pmu_platform_data *p_data =3D pmu->p_data; @@ -751,6 +955,37 @@ static const struct ddr_pmu_data ddr_pmu_data =3D { .id =3D DDR_PMU_V1, }; =20 +static const struct ddr_pmu_ops ddr_pmu_v2_ops =3D { + .enable_read_freerun_counter =3D ddr_pmu_v2_enable_read_freerun, + .enable_write_freerun_counter =3D ddr_pmu_v2_enable_write_freerun, + .clear_read_freerun_counter =3D ddr_pmu_v2_read_clear_freerun, + .clear_write_freerun_counter =3D ddr_pmu_v2_write_clear_freerun, + .pmu_overflow_handler =3D ddr_pmu_v2_overflow_hander, +}; + +static const struct ddr_pmu_platform_data odyssey_ddr_pmu_pdata =3D { + .counter_overflow_val =3D 0, + .counter_max_val =3D GENMASK_ULL(63, 0), + .ddrc_perf_cnt_base =3D ODY_DDRC_PERF_CNT_VALUE_BASE, + .ddrc_perf_cfg_base =3D ODY_DDRC_PERF_CFG_BASE, + .ddrc_perf_cnt_op_mode_ctrl =3D ODY_DDRC_PERF_CNT_OP_MODE_CTRL, + .ddrc_perf_cnt_start_op_ctrl =3D ODY_DDRC_PERF_CNT_START_OP_CTRL, + .ddrc_perf_cnt_end_op_ctrl =3D ODY_DDRC_PERF_CNT_END_OP_CTRL, + .ddrc_perf_cnt_end_status =3D ODY_DDRC_PERF_CNT_END_STATUS, + .ddrc_perf_cnt_freerun_en =3D 0, + .ddrc_perf_cnt_freerun_ctrl =3D ODY_DDRC_PERF_CNT_FREERUN_CTRL, + .ddrc_perf_cnt_freerun_clr =3D ODY_DDRC_PERF_CNT_FREERUN_CLR, + .ddrc_perf_cnt_value_wr_op =3D ODY_DDRC_PERF_CNT_VALUE_WR_OP, + .ddrc_perf_cnt_value_rd_op =3D ODY_DDRC_PERF_CNT_VALUE_RD_OP, + .ops =3D &ddr_pmu_v2_ops, +}; + +#ifdef CONFIG_ACPI +static const struct ddr_pmu_data ddr_pmu_v2_data =3D { + .id =3D DDR_PMU_V2, +}; +#endif + static int cn10k_ddr_perf_probe(struct platform_device *pdev) { const struct ddr_pmu_data *dev_data; @@ -803,6 +1038,21 @@ static int cn10k_ddr_perf_probe(struct platform_devic= e *pdev) .pmu_enable =3D cn10k_ddr_perf_pmu_enable, .pmu_disable =3D cn10k_ddr_perf_pmu_disable, }; + } else { + ddr_pmu->pmu =3D (struct pmu) { + .module =3D THIS_MODULE, + .capabilities =3D PERF_PMU_CAP_NO_EXCLUDE, + .task_ctx_nr =3D perf_invalid_context, + .attr_groups =3D odyssey_attr_groups, + .event_init =3D cn10k_ddr_perf_event_init, + .add =3D cn10k_ddr_perf_event_add, + .del =3D cn10k_ddr_perf_event_del, + .start =3D cn10k_ddr_perf_event_start, + .stop =3D cn10k_ddr_perf_event_stop, + .read =3D cn10k_ddr_perf_event_update, + }; + + ddr_pmu->p_data =3D &odyssey_ddr_pmu_pdata; } =20 /* Choose this cpu to collect perf data */ @@ -855,6 +1105,7 @@ MODULE_DEVICE_TABLE(of, cn10k_ddr_pmu_of_match); #ifdef CONFIG_ACPI static const struct acpi_device_id cn10k_ddr_pmu_acpi_match[] =3D { {"MRVL000A", (kernel_ulong_t)&ddr_pmu_data}, + {"MRVL000C", (kernel_ulong_t)&ddr_pmu_v2_data}, {}, }; MODULE_DEVICE_TABLE(acpi, cn10k_ddr_pmu_acpi_match); --=20 2.25.1 From nobody Fri Dec 19 16:53:31 2025 Received: from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com [67.231.156.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 0B4F81E2026 for ; Wed, 4 Sep 2024 18:26:53 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=67.231.156.173 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474415; cv=none; b=EaA+DLaJ2R1iJDiEHRlNaIfsezyPTXV4kFLqnye0ySrbM5IEOtvBHI8quHOh/U1x7qJy9G4jFnx/a/PAC4dSIos8NHJkKS6aCwwngV/S66wcGELbr43fCpLd8gzfQ+UJhnHdewGTHxMCrbdXaOBAGu3CJrvmmse2YUwyh6CrEKs= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474415; c=relaxed/simple; bh=MJyHgPPd/wjGZ3tiKzDqqqplfT19M+mdXcR2gOceW2w=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=PMY9M9g7bJ4iShWJFAs1LsmcKRnyRJ6IHJq+TksTeF+pVic3Ob0Df8msfAqxL7UA2S/RO4FvN6M7iYVX/UJBgzahleZDEWe78BjYb5lKnBO+FWJ3F88VP16zc51XzCX6HcdCC60Ko7pz9hI4IAydoaUPgHAElW5LfxxQa3Woh1o= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com; spf=pass smtp.mailfrom=marvell.com; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b=aF2u8rTZ; arc=none smtp.client-ip=67.231.156.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=marvell.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b="aF2u8rTZ" Received: from pps.filterd (m0045851.ppops.net [127.0.0.1]) by mx0b-0016f401.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 484HabO0000383; Wed, 4 Sep 2024 11:26:46 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pfpt0220; bh=s p78n+b6k2d5MDNfJaWsTjrFl1pMA438IW72EystjvU=; b=aF2u8rTZPshhPbCeq iwvuGWt5BvCz3l6zUtudDbpPLNLPnv9k8mNeCSpRUP0GvAFIG1CeqtQ0CEnvPAJp ZD0zVSk91GrJJDyVG9dVZ78XRKBKLbc76j6osyoLI5Si3OInHR6JgWaxy4BTZcrT JpP2ATiCo+wLUv8YgoeVb7Xdo22luf5BQ6JSiZDFIWyjy9UOcEROrrmSyDnxyAus utC/qVrkLZf5X+Zt4Qt9kLo7nsIgUgS2p9SwAPY9OmbKart339lf4oMhVpp0VEB1 /WL9+vTW20HH5px+HZFbbzMbK1eDvmN7GnG9xazZ99AafUy1aE8+S2ZbFH3StS91 EfCZQ== Received: from dc5-exch05.marvell.com ([199.233.59.128]) by mx0b-0016f401.pphosted.com (PPS) with ESMTPS id 41ev31r6gb-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 04 Sep 2024 11:26:46 -0700 (PDT) Received: from DC5-EXCH05.marvell.com (10.69.176.209) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.4; Wed, 4 Sep 2024 11:26:44 -0700 Received: from maili.marvell.com (10.69.176.80) by DC5-EXCH05.marvell.com (10.69.176.209) with Microsoft SMTP Server id 15.2.1544.4 via Frontend Transport; Wed, 4 Sep 2024 11:26:44 -0700 Received: from IPBU-BLR-SERVER1.marvell.com (IPBU-BLR-SERVER1.marvell.com [10.28.8.41]) by maili.marvell.com (Postfix) with ESMTP id 30B133F704F; Wed, 4 Sep 2024 11:26:41 -0700 (PDT) From: Gowthami Thiagarajan To: , , , CC: , , , Gowthami Thiagarajan Subject: [PATCH v7 5/6] perf/marvell : Refactor to extract platform data - no functional change Date: Wed, 4 Sep 2024 23:56:04 +0530 Message-ID: <20240904182605.953927-6-gthiagarajan@marvell.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240904182605.953927-1-gthiagarajan@marvell.com> References: <20240904182605.953927-1-gthiagarajan@marvell.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Proofpoint-ORIG-GUID: c2QlUyA3Po9ztrFq2HKDy7khOAw_mYuP X-Proofpoint-GUID: c2QlUyA3Po9ztrFq2HKDy7khOAw_mYuP X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-04_16,2024-09-04_01,2024-09-02_01 Content-Type: text/plain; charset="utf-8" This commit introduces a refactor to the Marvell TAD PMU driver specifically to add versioning to the existing driver. No functional changes are introduced; the behavior and performance of the driver remain unchanged. Signed-off-by: Gowthami Thiagarajan --- drivers/perf/marvell_cn10k_tad_pmu.c | 31 +++++++++++++++++++++++++--- 1 file changed, 28 insertions(+), 3 deletions(-) diff --git a/drivers/perf/marvell_cn10k_tad_pmu.c b/drivers/perf/marvell_cn= 10k_tad_pmu.c index 9e635f355470..15f9f67cb3bd 100644 --- a/drivers/perf/marvell_cn10k_tad_pmu.c +++ b/drivers/perf/marvell_cn10k_tad_pmu.c @@ -37,6 +37,14 @@ struct tad_pmu { DECLARE_BITMAP(counters_map, TAD_MAX_COUNTERS); }; =20 +enum mrvl_tad_pmu_version { + TAD_PMU_V1 =3D 1, +}; + +struct tad_pmu_data { + int id; +}; + static int tad_pmu_cpuhp_state; =20 static void tad_pmu_event_counter_read(struct perf_event *event) @@ -254,6 +262,7 @@ static const struct attribute_group *tad_pmu_attr_group= s[] =3D { =20 static int tad_pmu_probe(struct platform_device *pdev) { + const struct tad_pmu_data *dev_data; struct device *dev =3D &pdev->dev; struct tad_region *regions; struct tad_pmu *tad_pmu; @@ -261,6 +270,7 @@ static int tad_pmu_probe(struct platform_device *pdev) u32 tad_pmu_page_size; u32 tad_page_size; u32 tad_cnt; + int version; int i, ret; char *name; =20 @@ -270,6 +280,13 @@ static int tad_pmu_probe(struct platform_device *pdev) =20 platform_set_drvdata(pdev, tad_pmu); =20 + dev_data =3D device_get_match_data(&pdev->dev); + if (!dev_data) { + dev_err(&pdev->dev, "Error: No device match data found\n"); + return -ENODEV; + } + version =3D dev_data->id; + res =3D platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!res) { dev_err(&pdev->dev, "Mem resource not found\n"); @@ -319,7 +336,6 @@ static int tad_pmu_probe(struct platform_device *pdev) tad_pmu->pmu =3D (struct pmu) { =20 .module =3D THIS_MODULE, - .attr_groups =3D tad_pmu_attr_groups, .capabilities =3D PERF_PMU_CAP_NO_EXCLUDE | PERF_PMU_CAP_NO_INTERRUPT, .task_ctx_nr =3D perf_invalid_context, @@ -332,6 +348,9 @@ static int tad_pmu_probe(struct platform_device *pdev) .read =3D tad_pmu_event_counter_read, }; =20 + if (version =3D=3D TAD_PMU_V1) + tad_pmu->pmu.attr_groups =3D tad_pmu_attr_groups; + tad_pmu->cpu =3D raw_smp_processor_id(); =20 /* Register pmu instance for cpu hotplug */ @@ -360,16 +379,22 @@ static void tad_pmu_remove(struct platform_device *pd= ev) perf_pmu_unregister(&pmu->pmu); } =20 +#if defined(CONFIG_OF) || defined(CONFIG_ACPI) +static const struct tad_pmu_data tad_pmu_data =3D { + .id =3D TAD_PMU_V1, +}; +#endif + #ifdef CONFIG_OF static const struct of_device_id tad_pmu_of_match[] =3D { - { .compatible =3D "marvell,cn10k-tad-pmu", }, + { .compatible =3D "marvell,cn10k-tad-pmu", .data =3D &tad_pmu_data }, {}, }; #endif =20 #ifdef CONFIG_ACPI static const struct acpi_device_id tad_pmu_acpi_match[] =3D { - {"MRVL000B", 0}, + {"MRVL000B", (kernel_ulong_t)&tad_pmu_data}, {}, }; MODULE_DEVICE_TABLE(acpi, tad_pmu_acpi_match); --=20 2.25.1 From nobody Fri Dec 19 16:53:31 2025 Received: from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com [67.231.156.173]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 8B7B31E5002 for ; Wed, 4 Sep 2024 18:26:57 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=67.231.156.173 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474419; cv=none; b=PMq3AzbZo5eBkDsqeI7b2hh8Cj3U8/K6c5qnyr6Bfpbp7UOOASmnYb4+5MDuND7keigAQtESuloAJSvDOGoLRIoGr73EDkUTukYCwLl0CiyVzJ4gMHvqtYEAKS3vO7iAnYDe9fDuu5rok26kXSqecN374bOJJsWn50VlvVgayjA= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1725474419; c=relaxed/simple; bh=hR5viATnQMNf+Gc41v1mlpE5H8U0riYZ56AKkLDNy1s=; h=From:To:CC:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=sHMdL0H1J/5RXvmpTyJ5VDB7+07FEVmBZ+xB1gfrkd5T3QUBwSSjeSkVM13xdfR5c+okm3suyWHcFaGIEX4Xm6abWhNW15b0ccu0rI3oZOeN7uIXBentkY3SvwYV2cZK5WJ8h8KJTz3EfLk5/GdQg3xirdwk9JocW4Coo7/dpRs= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com; spf=pass smtp.mailfrom=marvell.com; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b=hx2YE7qa; arc=none smtp.client-ip=67.231.156.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=marvell.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=marvell.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=marvell.com header.i=@marvell.com header.b="hx2YE7qa" Received: from pps.filterd (m0045851.ppops.net [127.0.0.1]) by mx0b-0016f401.pphosted.com (8.18.1.2/8.18.1.2) with ESMTP id 484HZrxa031896; Wed, 4 Sep 2024 11:26:50 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pfpt0220; bh=R +5tDUt/Gwbuh3YnkRORAVssXlDTivKeHK5uU1tRuRQ=; b=hx2YE7qau7Ifwvfb8 MmNMfepog3r5PXtC6M7emXoEgZEWU8RQ4nbABHZeuo0w4emsb3gbWX8trarHJYZP KVb+ChY1PWb7WT8KIq8VSonup4C8RJonj2bpRWd7Q9g8NRpU5hCUC/XP21olvLDD sl19x0CjgbQ+EFOyqbvXisbAreQubECFyf0eCdtOUMzy0rJ4o56lCWcyjzvXfDQd 4ad9oovkUvidz4Q0COZGzqe/gkcpxS5n3VYJSelOPK2i/AkrNNCpaXHN3dEb1bbS CYHfhY7AL+5FJK0YJq2Exe8bpNpty0NCkCyySj/y17JWw3OXEbdZcAxGixylbXvf Q2E6g== Received: from dc6wp-exch02.marvell.com ([4.21.29.225]) by mx0b-0016f401.pphosted.com (PPS) with ESMTPS id 41ev31r6gg-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Wed, 04 Sep 2024 11:26:49 -0700 (PDT) Received: from DC6WP-EXCH02.marvell.com (10.76.176.209) by DC6WP-EXCH02.marvell.com (10.76.176.209) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id 15.2.1544.4; Wed, 4 Sep 2024 11:26:49 -0700 Received: from maili.marvell.com (10.69.176.80) by DC6WP-EXCH02.marvell.com (10.76.176.209) with Microsoft SMTP Server id 15.2.1544.4 via Frontend Transport; Wed, 4 Sep 2024 11:26:49 -0700 Received: from IPBU-BLR-SERVER1.marvell.com (IPBU-BLR-SERVER1.marvell.com [10.28.8.41]) by maili.marvell.com (Postfix) with ESMTP id 61B173F7044; Wed, 4 Sep 2024 11:26:46 -0700 (PDT) From: Gowthami Thiagarajan To: , , , CC: , , , Gowthami Thiagarajan Subject: [PATCH v7 6/6] perf/marvell : Odyssey LLC-TAD performance monitor support Date: Wed, 4 Sep 2024 23:56:05 +0530 Message-ID: <20240904182605.953927-7-gthiagarajan@marvell.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240904182605.953927-1-gthiagarajan@marvell.com> References: <20240904182605.953927-1-gthiagarajan@marvell.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Proofpoint-ORIG-GUID: MywRGd7AvAOw5kNHr-9ciQLqSDmuMisn X-Proofpoint-GUID: MywRGd7AvAOw5kNHr-9ciQLqSDmuMisn X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1039,Hydra:6.0.680,FMLib:17.12.60.29 definitions=2024-09-04_16,2024-09-04_01,2024-09-02_01 Content-Type: text/plain; charset="utf-8" Each TAD provides eight 64-bit counters for monitoring cache behavior.The driver always configures the same counter for all the TADs. The user would end up effectively reserving one of eight counters in every TAD to look across all TADs. The occurrences of events are aggregated and presented to the user at the end of running the workload. The driver does not provide a way for the user to partition TADs so that different TADs are used for different applications. The performance events reflect various internal or interface activities. By combining the values from multiple performance counters, cache performance can be measured in terms such as: cache miss rate, cache allocations, interface retry rate, internal resource occupancy, etc. Each supported counter's event and formatting information is exposed to sysfs at /sys/devices/tad/. Use perf tool stat command to measure the pmu events. For instance: perf stat -e tad_hit_ltg,tad_hit_dtg Signed-off-by: Gowthami Thiagarajan --- Documentation/admin-guide/perf/index.rst | 1 + .../admin-guide/perf/mrvl-odyssey-tad-pmu.rst | 37 +++++++++++++++++++ drivers/perf/marvell_cn10k_tad_pmu.c | 35 ++++++++++++++++++ 3 files changed, 73 insertions(+) create mode 100644 Documentation/admin-guide/perf/mrvl-odyssey-tad-pmu.rst diff --git a/Documentation/admin-guide/perf/index.rst b/Documentation/admin= -guide/perf/index.rst index d673ccfea903..d8e983e33ca7 100644 --- a/Documentation/admin-guide/perf/index.rst +++ b/Documentation/admin-guide/perf/index.rst @@ -15,6 +15,7 @@ Performance monitor support qcom_l3_pmu starfive_starlink_pmu mrvl-odyssey-ddr-pmu + mrvl-odyssey-tad-pmu arm-ccn arm-cmn xgene-pmu diff --git a/Documentation/admin-guide/perf/mrvl-odyssey-tad-pmu.rst b/Docu= mentation/admin-guide/perf/mrvl-odyssey-tad-pmu.rst new file mode 100644 index 000000000000..ad1975b14087 --- /dev/null +++ b/Documentation/admin-guide/perf/mrvl-odyssey-tad-pmu.rst @@ -0,0 +1,37 @@ +=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D +Marvell Odyssey LLC-TAD Performance Monitoring Unit (PMU UNCORE) +=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D + +Each TAD provides eight 64-bit counters for monitoring +cache behavior.The driver always configures the same counter for +all the TADs. The user would end up effectively reserving one of +eight counters in every TAD to look across all TADs. +The occurrences of events are aggregated and presented to the user +at the end of running the workload. The driver does not provide a +way for the user to partition TADs so that different TADs are used for +different applications. + +The performance events reflect various internal or interface activities. +By combining the values from multiple performance counters, cache +performance can be measured in terms such as: cache miss rate, cache +allocations, interface retry rate, internal resource occupancy, etc. + +The PMU driver exposes the available events and format options under sysfs= :: + + /sys/bus/event_source/devices/tad/events/ + /sys/bus/event_source/devices/tad/format/ + +Examples:: + + $ perf list | grep tad + tad/tad_alloc_any/ [Kernel PMU eve= nt] + tad/tad_alloc_dtg/ [Kernel PMU eve= nt] + tad/tad_alloc_ltg/ [Kernel PMU eve= nt] + tad/tad_hit_any/ [Kernel PMU eve= nt] + tad/tad_hit_dtg/ [Kernel PMU eve= nt] + tad/tad_hit_ltg/ [Kernel PMU eve= nt] + tad/tad_req_msh_in_exlmn/ [Kernel PMU eve= nt] + tad/tad_tag_rd/ [Kernel PMU eve= nt] + tad/tad_tot_cycle/ [Kernel PMU eve= nt] + + $ perf stat -e tad_alloc_dtg,tad_alloc_ltg,tad_alloc_any,tad_hit_dtg,ta= d_hit_ltg,tad_hit_any,tad_tag_rd diff --git a/drivers/perf/marvell_cn10k_tad_pmu.c b/drivers/perf/marvell_cn= 10k_tad_pmu.c index 15f9f67cb3bd..29976b435417 100644 --- a/drivers/perf/marvell_cn10k_tad_pmu.c +++ b/drivers/perf/marvell_cn10k_tad_pmu.c @@ -39,6 +39,7 @@ struct tad_pmu { =20 enum mrvl_tad_pmu_version { TAD_PMU_V1 =3D 1, + TAD_PMU_V2, }; =20 struct tad_pmu_data { @@ -222,6 +223,24 @@ static const struct attribute_group tad_pmu_events_att= r_group =3D { .attrs =3D tad_pmu_event_attrs, }; =20 +static struct attribute *ody_tad_pmu_event_attrs[] =3D { + TAD_PMU_EVENT_ATTR(tad_req_msh_in_exlmn, 0x3), + TAD_PMU_EVENT_ATTR(tad_alloc_dtg, 0x1a), + TAD_PMU_EVENT_ATTR(tad_alloc_ltg, 0x1b), + TAD_PMU_EVENT_ATTR(tad_alloc_any, 0x1c), + TAD_PMU_EVENT_ATTR(tad_hit_dtg, 0x1d), + TAD_PMU_EVENT_ATTR(tad_hit_ltg, 0x1e), + TAD_PMU_EVENT_ATTR(tad_hit_any, 0x1f), + TAD_PMU_EVENT_ATTR(tad_tag_rd, 0x20), + TAD_PMU_EVENT_ATTR(tad_tot_cycle, 0xFF), + NULL +}; + +static const struct attribute_group ody_tad_pmu_events_attr_group =3D { + .name =3D "events", + .attrs =3D ody_tad_pmu_event_attrs, +}; + PMU_FORMAT_ATTR(event, "config:0-7"); =20 static struct attribute *tad_pmu_format_attrs[] =3D { @@ -260,6 +279,13 @@ static const struct attribute_group *tad_pmu_attr_grou= ps[] =3D { NULL }; =20 +static const struct attribute_group *ody_tad_pmu_attr_groups[] =3D { + &ody_tad_pmu_events_attr_group, + &tad_pmu_format_attr_group, + &tad_pmu_cpumask_attr_group, + NULL +}; + static int tad_pmu_probe(struct platform_device *pdev) { const struct tad_pmu_data *dev_data; @@ -350,6 +376,8 @@ static int tad_pmu_probe(struct platform_device *pdev) =20 if (version =3D=3D TAD_PMU_V1) tad_pmu->pmu.attr_groups =3D tad_pmu_attr_groups; + else + tad_pmu->pmu.attr_groups =3D ody_tad_pmu_attr_groups; =20 tad_pmu->cpu =3D raw_smp_processor_id(); =20 @@ -385,6 +413,12 @@ static const struct tad_pmu_data tad_pmu_data =3D { }; #endif =20 +#ifdef CONFIG_ACPI +static const struct tad_pmu_data tad_pmu_v2_data =3D { + .id =3D TAD_PMU_V2, +}; +#endif + #ifdef CONFIG_OF static const struct of_device_id tad_pmu_of_match[] =3D { { .compatible =3D "marvell,cn10k-tad-pmu", .data =3D &tad_pmu_data }, @@ -395,6 +429,7 @@ static const struct of_device_id tad_pmu_of_match[] =3D= { #ifdef CONFIG_ACPI static const struct acpi_device_id tad_pmu_acpi_match[] =3D { {"MRVL000B", (kernel_ulong_t)&tad_pmu_data}, + {"MRVL000D", (kernel_ulong_t)&tad_pmu_v2_data}, {}, }; MODULE_DEVICE_TABLE(acpi, tad_pmu_acpi_match); --=20 2.25.1