From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-9.0 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_PATCH,MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS,URIBL_BLOCKED, USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id CD73CC64EB4 for ; Sat, 1 Dec 2018 00:58:07 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 869ED20673 for ; Sat, 1 Dec 2018 00:58:07 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 869ED20673 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=anholt.net Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726736AbeLAMJR (ORCPT ); Sat, 1 Dec 2018 07:09:17 -0500 Received: from anholt.net ([50.246.234.109]:42748 "EHLO anholt.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726014AbeLAMJQ (ORCPT ); Sat, 1 Dec 2018 07:09:16 -0500 Received: from localhost (localhost [127.0.0.1]) by anholt.net (Postfix) with ESMTP id 2958710A156D; Fri, 30 Nov 2018 16:58:04 -0800 (PST) X-Virus-Scanned: Debian amavisd-new at anholt.net Received: from anholt.net ([127.0.0.1]) by localhost (kingsolver.anholt.net [127.0.0.1]) (amavisd-new, port 10024) with LMTP id w7ud59fdKjYI; Fri, 30 Nov 2018 16:58:02 -0800 (PST) Received: from eliezer.anholt.net (localhost [127.0.0.1]) by anholt.net (Postfix) with ESMTP id 7DE2410A1521; Fri, 30 Nov 2018 16:58:00 -0800 (PST) Received: by eliezer.anholt.net (Postfix, from userid 1000) id 73E4C2FE36F0; Fri, 30 Nov 2018 16:57:59 -0800 (PST) From: Eric Anholt To: dri-devel@lists.freedesktop.org Cc: linux-kernel@vger.kernel.org, Dave Emett , Thomas Spurden , Eric Anholt Subject: [PATCH 5/6] drm/v3d: Add more tracepoints for V3D GPU rendering. Date: Fri, 30 Nov 2018 16:57:58 -0800 Message-Id: <20181201005759.28093-5-eric@anholt.net> X-Mailer: git-send-email 2.20.0.rc1 In-Reply-To: <20181201005759.28093-1-eric@anholt.net> References: <20181201005759.28093-1-eric@anholt.net> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The core scheduler tells us when the job is pushed to the scheduler's queue, and I had the job_run functions saying when they actually queue the job to the hardware. By adding tracepoints for the very top of the ioctls and the IRQs signaling job completion, "perf record -a -e v3d:.\* -e gpu_scheduler:.\* ; perf script" gets you a pretty decent timeline. Signed-off-by: Eric Anholt --- drivers/gpu/drm/v3d/v3d_gem.c | 4 ++ drivers/gpu/drm/v3d/v3d_irq.c | 19 +++++- drivers/gpu/drm/v3d/v3d_trace.h | 101 ++++++++++++++++++++++++++++++++ 3 files changed, 121 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c index 0bd6892e3044..2f82e2724b1f 100644 --- a/drivers/gpu/drm/v3d/v3d_gem.c +++ b/drivers/gpu/drm/v3d/v3d_gem.c @@ -484,6 +484,8 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data, struct drm_syncobj *sync_out; int ret = 0; + trace_v3d_submit_cl_ioctl(&v3d->drm, args->rcl_start, args->rcl_end); + if (args->pad != 0) { DRM_INFO("pad must be zero: %d\n", args->pad); return -EINVAL; @@ -611,6 +613,8 @@ v3d_submit_tfu_ioctl(struct drm_device *dev, void *data, int ret = 0; int bo_count; + trace_v3d_submit_tfu_ioctl(&v3d->drm, args->iia); + job = kcalloc(1, sizeof(*job), GFP_KERNEL); if (!job) return -ENOMEM; diff --git a/drivers/gpu/drm/v3d/v3d_irq.c b/drivers/gpu/drm/v3d/v3d_irq.c index dd7a7b0bd5a1..69338da70ddc 100644 --- a/drivers/gpu/drm/v3d/v3d_irq.c +++ b/drivers/gpu/drm/v3d/v3d_irq.c @@ -15,6 +15,7 @@ #include "v3d_drv.h" #include "v3d_regs.h" +#include "v3d_trace.h" #define V3D_CORE_IRQS ((u32)(V3D_INT_OUTOMEM | \ V3D_INT_FLDONE | \ @@ -88,12 +89,20 @@ v3d_irq(int irq, void *arg) } if (intsts & V3D_INT_FLDONE) { - dma_fence_signal(v3d->bin_job->bin.done_fence); + struct v3d_fence *fence = + to_v3d_fence(v3d->bin_job->bin.done_fence); + + trace_v3d_bcl_irq(&v3d->drm, fence->seqno); + dma_fence_signal(&fence->base); status = IRQ_HANDLED; } if (intsts & V3D_INT_FRDONE) { - dma_fence_signal(v3d->render_job->render.done_fence); + struct v3d_fence *fence = + to_v3d_fence(v3d->render_job->render.done_fence); + + trace_v3d_rcl_irq(&v3d->drm, fence->seqno); + dma_fence_signal(&fence->base); status = IRQ_HANDLED; } @@ -119,7 +128,11 @@ v3d_hub_irq(int irq, void *arg) V3D_WRITE(V3D_HUB_INT_CLR, intsts); if (intsts & V3D_HUB_INT_TFUC) { - dma_fence_signal(v3d->tfu_job->done_fence); + struct v3d_fence *fence = + to_v3d_fence(v3d->tfu_job->done_fence); + + trace_v3d_tfu_irq(&v3d->drm, fence->seqno); + dma_fence_signal(&fence->base); status = IRQ_HANDLED; } diff --git a/drivers/gpu/drm/v3d/v3d_trace.h b/drivers/gpu/drm/v3d/v3d_trace.h index f54ed9cd3444..edd984afa33f 100644 --- a/drivers/gpu/drm/v3d/v3d_trace.h +++ b/drivers/gpu/drm/v3d/v3d_trace.h @@ -12,6 +12,28 @@ #define TRACE_SYSTEM v3d #define TRACE_INCLUDE_FILE v3d_trace +TRACE_EVENT(v3d_submit_cl_ioctl, + TP_PROTO(struct drm_device *dev, u32 ct1qba, u32 ct1qea), + TP_ARGS(dev, ct1qba, ct1qea), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u32, ct1qba) + __field(u32, ct1qea) + ), + + TP_fast_assign( + __entry->dev = dev->primary->index; + __entry->ct1qba = ct1qba; + __entry->ct1qea = ct1qea; + ), + + TP_printk("dev=%u, RCL 0x%08x..0x%08x", + __entry->dev, + __entry->ct1qba, + __entry->ct1qea) +); + TRACE_EVENT(v3d_submit_cl, TP_PROTO(struct drm_device *dev, bool is_render, uint64_t seqno, @@ -42,6 +64,85 @@ TRACE_EVENT(v3d_submit_cl, __entry->ctnqea) ); +TRACE_EVENT(v3d_bcl_irq, + TP_PROTO(struct drm_device *dev, + uint64_t seqno), + TP_ARGS(dev, seqno), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u64, seqno) + ), + + TP_fast_assign( + __entry->dev = dev->primary->index; + __entry->seqno = seqno; + ), + + TP_printk("dev=%u, seqno=%llu", + __entry->dev, + __entry->seqno) +); + +TRACE_EVENT(v3d_rcl_irq, + TP_PROTO(struct drm_device *dev, + uint64_t seqno), + TP_ARGS(dev, seqno), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u64, seqno) + ), + + TP_fast_assign( + __entry->dev = dev->primary->index; + __entry->seqno = seqno; + ), + + TP_printk("dev=%u, seqno=%llu", + __entry->dev, + __entry->seqno) +); + +TRACE_EVENT(v3d_tfu_irq, + TP_PROTO(struct drm_device *dev, + uint64_t seqno), + TP_ARGS(dev, seqno), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u64, seqno) + ), + + TP_fast_assign( + __entry->dev = dev->primary->index; + __entry->seqno = seqno; + ), + + TP_printk("dev=%u, seqno=%llu", + __entry->dev, + __entry->seqno) +); + +TRACE_EVENT(v3d_submit_tfu_ioctl, + TP_PROTO(struct drm_device *dev, u32 iia), + TP_ARGS(dev, iia), + + TP_STRUCT__entry( + __field(u32, dev) + __field(u32, iia) + ), + + TP_fast_assign( + __entry->dev = dev->primary->index; + __entry->iia = iia; + ), + + TP_printk("dev=%u, IIA 0x%08x", + __entry->dev, + __entry->iia) +); + TRACE_EVENT(v3d_submit_tfu, TP_PROTO(struct drm_device *dev, uint64_t seqno), -- 2.20.0.rc1