From patchwork Mon Aug 19 08:02:22 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mary Guillemard X-Patchwork-Id: 13768046 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (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 48F29C52D7C for ; Mon, 19 Aug 2024 08:02:57 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id CD55610E1E4; Mon, 19 Aug 2024 08:02:56 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (1024-bit key; unprotected) header.d=collabora.com header.i=mary.guillemard@collabora.com header.b="El8OcJ7t"; dkim-atps=neutral Received: from sender4-pp-f112.zoho.com (sender4-pp-f112.zoho.com [136.143.188.112]) by gabe.freedesktop.org (Postfix) with ESMTPS id BAE9710E1E4 for ; Mon, 19 Aug 2024 08:02:55 +0000 (UTC) ARC-Seal: i=1; a=rsa-sha256; t=1724054564; cv=none; d=zohomail.com; s=zohoarc; b=d7dXlTJLid7xsC+QZFjdix6aydqyKVJhIXfvW+OVcBKK0X61EEA6xxJm2n1sa1FEMHsbaLGKBRFfgcj/RYu9ET7JmQVFyHdyGBRdSVsi0litOE4vIyJzmpzK/GLf9wzUJ6fFDHoIeALiF7Tvvov6UZMphSEKP5RhP1eyYIBpd4g= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1724054564; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:MIME-Version:Message-ID:References:Subject:Subject:To:To:Message-Id:Reply-To; bh=FMGdiUWlsSaMtAviONF1MB/nVlk0jItnN0mLRSQtc5w=; b=aSTcXedORF9U9M4F3hMqqyKOqedGnVgwZu5MU/8a0g0MWwSGjXadUIA64Rz1EmaE39/NEbKeJhVcXJMCHKsJST7RenivSOv2FOxKkl4HsOkpfLAOcJfonjh9xZOATrReQdraaIRBOP5atxhP6Y41BlCIintIZU79yeQJwvuiOlI= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=collabora.com; spf=pass smtp.mailfrom=mary.guillemard@collabora.com; dmarc=pass header.from= DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; t=1724054564; s=zohomail; d=collabora.com; i=mary.guillemard@collabora.com; h=From:From:To:To:Cc:Cc:Subject:Subject:Date:Date:Message-ID:In-Reply-To:References:MIME-Version:Content-Transfer-Encoding:Message-Id:Reply-To; bh=FMGdiUWlsSaMtAviONF1MB/nVlk0jItnN0mLRSQtc5w=; b=El8OcJ7tk8OnL2YW98r6HBraCrzS9F/j9/V6+51uXBMa/XpRrO6yz5WeZUSNRBJT xspE/I83XnN1AyK2AsLt/PNuP8jDgl6L/cMqn7lFjsfXP6UKrq/UTuaGZ9CFd1zfsJd z8+7grBPxxElc2KE1+LIOfmVLOsn4njbCyh2nQmo= Received: by mx.zohomail.com with SMTPS id 1724054563190586.4427888975266; Mon, 19 Aug 2024 01:02:43 -0700 (PDT) From: Mary Guillemard To: linux-kernel@vger.kernel.org Cc: dri-devel@lists.freedesktop.org, kernel@collabora.com, Mary Guillemard , Boris Brezillon , Rob Herring , Steven Price , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter Subject: [PATCH v3 1/2] drm/panfrost: Add SYSTEM_TIMESTAMP and SYSTEM_TIMESTAMP_FREQUENCY parameters Date: Mon, 19 Aug 2024 10:02:22 +0200 Message-ID: <20240819080224.24914-2-mary.guillemard@collabora.com> X-Mailer: git-send-email 2.46.0 In-Reply-To: <20240819080224.24914-1-mary.guillemard@collabora.com> References: <20240819080224.24914-1-mary.guillemard@collabora.com> MIME-Version: 1.0 X-ZohoMailClient: External X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" Expose system timestamp and frequency supported by the GPU. Mali uses an external timer as GPU system time. On ARM, this is wired to the generic arch timer so we wire cntfrq_el0 as device frequency. This new uAPI will be used in Mesa to implement timestamp queries and VK_KHR_calibrated_timestamps. v2: - Rewrote to use GPU timestamp register - Add missing include for arch_timer_get_cntfrq - Rework commit message v3: - Move panfrost_cycle_counter_get and panfrost_cycle_counter_put to panfrost_ioctl_query_timestamp - Handle possible overflow in panfrost_timestamp_read Signed-off-by: Mary Guillemard Tested-by: Heiko Stuebner Reviewed-by: Steven Price --- drivers/gpu/drm/panfrost/panfrost_drv.c | 37 ++++++++++++++++++++++++ drivers/gpu/drm/panfrost/panfrost_gpu.c | 12 ++++++++ drivers/gpu/drm/panfrost/panfrost_gpu.h | 1 + drivers/gpu/drm/panfrost/panfrost_regs.h | 2 ++ include/uapi/drm/panfrost_drm.h | 2 ++ 5 files changed, 54 insertions(+) diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c index 671eed4ad890..790c4ad31143 100644 --- a/drivers/gpu/drm/panfrost/panfrost_drv.c +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c @@ -3,6 +3,10 @@ /* Copyright 2019 Linaro, Ltd., Rob Herring */ /* Copyright 2019 Collabora ltd. */ +#ifdef CONFIG_ARM_ARCH_TIMER +#include +#endif + #include #include #include @@ -24,10 +28,28 @@ static bool unstable_ioctls; module_param_unsafe(unstable_ioctls, bool, 0600); +static int panfrost_ioctl_query_timestamp(struct panfrost_device *pfdev, + u64 *arg) +{ + int ret; + + ret = pm_runtime_resume_and_get(pfdev->dev); + if (ret) + return ret; + + panfrost_cycle_counter_get(pfdev); + *arg = panfrost_timestamp_read(pfdev); + panfrost_cycle_counter_put(pfdev); + + pm_runtime_put(pfdev->dev); + return 0; +} + static int panfrost_ioctl_get_param(struct drm_device *ddev, void *data, struct drm_file *file) { struct drm_panfrost_get_param *param = data; struct panfrost_device *pfdev = ddev->dev_private; + int ret; if (param->pad != 0) return -EINVAL; @@ -69,6 +91,21 @@ static int panfrost_ioctl_get_param(struct drm_device *ddev, void *data, struct PANFROST_FEATURE_ARRAY(JS_FEATURES, js_features, 15); PANFROST_FEATURE(NR_CORE_GROUPS, nr_core_groups); PANFROST_FEATURE(THREAD_TLS_ALLOC, thread_tls_alloc); + + case DRM_PANFROST_PARAM_SYSTEM_TIMESTAMP: + ret = panfrost_ioctl_query_timestamp(pfdev, ¶m->value); + if (ret) + return ret; + break; + + case DRM_PANFROST_PARAM_SYSTEM_TIMESTAMP_FREQUENCY: +#ifdef CONFIG_ARM_ARCH_TIMER + param->value = arch_timer_get_cntfrq(); +#else + param->value = 0; +#endif + break; + default: return -EINVAL; } diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c index fd8e44992184..f19f918e2330 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gpu.c +++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c @@ -380,6 +380,18 @@ unsigned long long panfrost_cycle_counter_read(struct panfrost_device *pfdev) return ((u64)hi << 32) | lo; } +unsigned long long panfrost_timestamp_read(struct panfrost_device *pfdev) +{ + u32 hi, lo; + + do { + hi = gpu_read(pfdev, GPU_TIMESTAMP_HI); + lo = gpu_read(pfdev, GPU_TIMESTAMP_LO); + } while (hi != gpu_read(pfdev, GPU_TIMESTAMP_HI)); + + return ((u64)hi << 32) | lo; +} + static u64 panfrost_get_core_mask(struct panfrost_device *pfdev) { u64 core_mask; diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.h b/drivers/gpu/drm/panfrost/panfrost_gpu.h index d841b86504ea..b4fef11211d5 100644 --- a/drivers/gpu/drm/panfrost/panfrost_gpu.h +++ b/drivers/gpu/drm/panfrost/panfrost_gpu.h @@ -20,6 +20,7 @@ void panfrost_gpu_suspend_irq(struct panfrost_device *pfdev); void panfrost_cycle_counter_get(struct panfrost_device *pfdev); void panfrost_cycle_counter_put(struct panfrost_device *pfdev); unsigned long long panfrost_cycle_counter_read(struct panfrost_device *pfdev); +unsigned long long panfrost_timestamp_read(struct panfrost_device *pfdev); void panfrost_gpu_amlogic_quirk(struct panfrost_device *pfdev); diff --git a/drivers/gpu/drm/panfrost/panfrost_regs.h b/drivers/gpu/drm/panfrost/panfrost_regs.h index c25743b05c55..c7bba476ab3f 100644 --- a/drivers/gpu/drm/panfrost/panfrost_regs.h +++ b/drivers/gpu/drm/panfrost/panfrost_regs.h @@ -78,6 +78,8 @@ #define GPU_CYCLE_COUNT_LO 0x90 #define GPU_CYCLE_COUNT_HI 0x94 +#define GPU_TIMESTAMP_LO 0x98 +#define GPU_TIMESTAMP_HI 0x9C #define GPU_THREAD_MAX_THREADS 0x0A0 /* (RO) Maximum number of threads per core */ #define GPU_THREAD_MAX_WORKGROUP_SIZE 0x0A4 /* (RO) Maximum workgroup size */ diff --git a/include/uapi/drm/panfrost_drm.h b/include/uapi/drm/panfrost_drm.h index 9f231d40a146..52b050e2b660 100644 --- a/include/uapi/drm/panfrost_drm.h +++ b/include/uapi/drm/panfrost_drm.h @@ -172,6 +172,8 @@ enum drm_panfrost_param { DRM_PANFROST_PARAM_NR_CORE_GROUPS, DRM_PANFROST_PARAM_THREAD_TLS_ALLOC, DRM_PANFROST_PARAM_AFBC_FEATURES, + DRM_PANFROST_PARAM_SYSTEM_TIMESTAMP, + DRM_PANFROST_PARAM_SYSTEM_TIMESTAMP_FREQUENCY, }; struct drm_panfrost_get_param { From patchwork Mon Aug 19 08:02:23 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mary Guillemard X-Patchwork-Id: 13768047 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 gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (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 9BDEFC5320E for ; Mon, 19 Aug 2024 08:02:59 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 21A9410E1E5; Mon, 19 Aug 2024 08:02:59 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (1024-bit key; unprotected) header.d=collabora.com header.i=mary.guillemard@collabora.com header.b="jT4O58UY"; dkim-atps=neutral Received: from sender4-pp-f112.zoho.com (sender4-pp-f112.zoho.com [136.143.188.112]) by gabe.freedesktop.org (Postfix) with ESMTPS id 6B62310E1E5 for ; Mon, 19 Aug 2024 08:02:57 +0000 (UTC) ARC-Seal: i=1; a=rsa-sha256; t=1724054568; cv=none; d=zohomail.com; s=zohoarc; b=M2WVo+Ly3Y9i7hmUKWlpWpanxR1MtfQEiToDtqT6p3zhWSCfCRaaAUBHQTlvN8Akp5yAG9kWRJ47FP2uq0yNCiLoM2SIbPrmTAm2/slrXzqrhRZkkzuKTbW3qd4BX5weB/jEDGzrf2vs+rvyxrYHX+5M+YwY4WB/1jx+ZdIqD/Y= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1724054568; h=Content-Transfer-Encoding:Cc:Cc:Date:Date:From:From:In-Reply-To:MIME-Version:Message-ID:References:Subject:Subject:To:To:Message-Id:Reply-To; bh=WM+wqGp9qLzx1kIn1dgpyGNYvgrZqELl8IsnsJ4nghA=; b=i3wY5IXaM9777plSt9Pkeop1MYo24Lhb023yXzZqMUeH5j/AkLGmWUFoF3OX1lPzdT0hf7vR8dkhILzole4WAQ6WB2GXqJIebFk0Mo+Ww05rXPkVyi4pNlvUbz0SfrdqvxMMV4eGe74KOdumSZquKEO9JIqjBrYyqwNhP6N1JFs= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass header.i=collabora.com; spf=pass smtp.mailfrom=mary.guillemard@collabora.com; dmarc=pass header.from= DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; t=1724054568; s=zohomail; d=collabora.com; i=mary.guillemard@collabora.com; h=From:From:To:To:Cc:Cc:Subject:Subject:Date:Date:Message-ID:In-Reply-To:References:MIME-Version:Content-Transfer-Encoding:Message-Id:Reply-To; bh=WM+wqGp9qLzx1kIn1dgpyGNYvgrZqELl8IsnsJ4nghA=; b=jT4O58UYzlEkd6sLATagyUvvAE1kw1LylLw3yBsWf4zVeVBD+ZitzvW08HoL8Qem 27dtV19lTyh+LBpJSVFCFPM/HNHAwv3gfkuR/5BmL9an6hnyAFq7Te/LqqXHKcfxFez Mfr7sXpVrNcRcKpsrq8WJceyugNz/y+wrq4pm2SE= Received: by mx.zohomail.com with SMTPS id 1724054566910393.5531239813379; Mon, 19 Aug 2024 01:02:46 -0700 (PDT) From: Mary Guillemard To: linux-kernel@vger.kernel.org Cc: dri-devel@lists.freedesktop.org, kernel@collabora.com, Mary Guillemard , Steven Price , Boris Brezillon , Rob Herring , Maarten Lankhorst , Maxime Ripard , Thomas Zimmermann , David Airlie , Daniel Vetter Subject: [PATCH v3 2/2] drm/panfrost: Add cycle counter job requirement Date: Mon, 19 Aug 2024 10:02:23 +0200 Message-ID: <20240819080224.24914-3-mary.guillemard@collabora.com> X-Mailer: git-send-email 2.46.0 In-Reply-To: <20240819080224.24914-1-mary.guillemard@collabora.com> References: <20240819080224.24914-1-mary.guillemard@collabora.com> MIME-Version: 1.0 X-ZohoMailClient: External X-BeenThere: dri-devel@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Direct Rendering Infrastructure - Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" Extend the uAPI with a new job requirement flag for cycle counters. This requirement is used by userland to indicate that a job requires cycle counters or system timestamp to be propagated. (for use with write value timestamp jobs) We cannot enable cycle counters unconditionally as this would result in an increase of GPU power consumption. As a result, they should be left off unless required by the application. If a job requires cycle counters or system timestamps propagation, we must enable cycle counting before issuing a job and disable it right after the job completes. Since this extends the uAPI and because userland needs a way to advertise features like VK_KHR_shader_clock conditionally, we bumps the driver minor version. v2: - Rework commit message - Squash uAPI changes and implementation in this commit - Simplify changes based on Steven Price comments v3: - Add Steven Price r-b - Fix a codestyle issue Signed-off-by: Mary Guillemard Reviewed-by: Steven Price Tested-by: Heiko Stuebner --- drivers/gpu/drm/panfrost/panfrost_drv.c | 8 +++++-- drivers/gpu/drm/panfrost/panfrost_job.c | 28 +++++++++++++++---------- include/uapi/drm/panfrost_drm.h | 1 + 3 files changed, 24 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/panfrost/panfrost_drv.c b/drivers/gpu/drm/panfrost/panfrost_drv.c index 790c4ad31143..04d615df5259 100644 --- a/drivers/gpu/drm/panfrost/panfrost_drv.c +++ b/drivers/gpu/drm/panfrost/panfrost_drv.c @@ -25,6 +25,8 @@ #include "panfrost_gpu.h" #include "panfrost_perfcnt.h" +#define JOB_REQUIREMENTS (PANFROST_JD_REQ_FS | PANFROST_JD_REQ_CYCLE_COUNT) + static bool unstable_ioctls; module_param_unsafe(unstable_ioctls, bool, 0600); @@ -282,7 +284,7 @@ static int panfrost_ioctl_submit(struct drm_device *dev, void *data, if (!args->jc) return -EINVAL; - if (args->requirements && args->requirements != PANFROST_JD_REQ_FS) + if (args->requirements & ~JOB_REQUIREMENTS) return -EINVAL; if (args->out_sync > 0) { @@ -621,6 +623,8 @@ static const struct file_operations panfrost_drm_driver_fops = { * - 1.0 - initial interface * - 1.1 - adds HEAP and NOEXEC flags for CREATE_BO * - 1.2 - adds AFBC_FEATURES query + * - 1.3 - adds JD_REQ_CYCLE_COUNT job requirement for SUBMIT + * - adds SYSTEM_TIMESTAMP and SYSTEM_TIMESTAMP_FREQUENCY queries */ static const struct drm_driver panfrost_drm_driver = { .driver_features = DRIVER_RENDER | DRIVER_GEM | DRIVER_SYNCOBJ, @@ -634,7 +638,7 @@ static const struct drm_driver panfrost_drm_driver = { .desc = "panfrost DRM", .date = "20180908", .major = 1, - .minor = 2, + .minor = 3, .gem_create_object = panfrost_gem_create_object, .gem_prime_import_sg_table = panfrost_gem_prime_import_sg_table, diff --git a/drivers/gpu/drm/panfrost/panfrost_job.c b/drivers/gpu/drm/panfrost/panfrost_job.c index df49d37d0e7e..3ad131eb6657 100644 --- a/drivers/gpu/drm/panfrost/panfrost_job.c +++ b/drivers/gpu/drm/panfrost/panfrost_job.c @@ -159,16 +159,17 @@ panfrost_dequeue_job(struct panfrost_device *pfdev, int slot) struct panfrost_job *job = pfdev->jobs[slot][0]; WARN_ON(!job); - if (job->is_profiled) { - if (job->engine_usage) { - job->engine_usage->elapsed_ns[slot] += - ktime_to_ns(ktime_sub(ktime_get(), job->start_time)); - job->engine_usage->cycles[slot] += - panfrost_cycle_counter_read(pfdev) - job->start_cycles; - } - panfrost_cycle_counter_put(job->pfdev); + + if (job->is_profiled && job->engine_usage) { + job->engine_usage->elapsed_ns[slot] += + ktime_to_ns(ktime_sub(ktime_get(), job->start_time)); + job->engine_usage->cycles[slot] += + panfrost_cycle_counter_read(pfdev) - job->start_cycles; } + if (job->requirements & PANFROST_JD_REQ_CYCLE_COUNT || job->is_profiled) + panfrost_cycle_counter_put(pfdev); + pfdev->jobs[slot][0] = pfdev->jobs[slot][1]; pfdev->jobs[slot][1] = NULL; @@ -243,9 +244,13 @@ static void panfrost_job_hw_submit(struct panfrost_job *job, int js) subslot = panfrost_enqueue_job(pfdev, js, job); /* Don't queue the job if a reset is in progress */ if (!atomic_read(&pfdev->reset.pending)) { - if (pfdev->profile_mode) { + job->is_profiled = pfdev->profile_mode; + + if (job->requirements & PANFROST_JD_REQ_CYCLE_COUNT || + job->is_profiled) panfrost_cycle_counter_get(pfdev); - job->is_profiled = true; + + if (job->is_profiled) { job->start_time = ktime_get(); job->start_cycles = panfrost_cycle_counter_read(pfdev); } @@ -693,7 +698,8 @@ panfrost_reset(struct panfrost_device *pfdev, spin_lock(&pfdev->js->job_lock); for (i = 0; i < NUM_JOB_SLOTS; i++) { for (j = 0; j < ARRAY_SIZE(pfdev->jobs[0]) && pfdev->jobs[i][j]; j++) { - if (pfdev->jobs[i][j]->is_profiled) + if (pfdev->jobs[i][j]->requirements & PANFROST_JD_REQ_CYCLE_COUNT || + pfdev->jobs[i][j]->is_profiled) panfrost_cycle_counter_put(pfdev->jobs[i][j]->pfdev); pm_runtime_put_noidle(pfdev->dev); panfrost_devfreq_record_idle(&pfdev->pfdevfreq); diff --git a/include/uapi/drm/panfrost_drm.h b/include/uapi/drm/panfrost_drm.h index 52b050e2b660..568724be6628 100644 --- a/include/uapi/drm/panfrost_drm.h +++ b/include/uapi/drm/panfrost_drm.h @@ -40,6 +40,7 @@ extern "C" { #define DRM_IOCTL_PANFROST_PERFCNT_DUMP DRM_IOW(DRM_COMMAND_BASE + DRM_PANFROST_PERFCNT_DUMP, struct drm_panfrost_perfcnt_dump) #define PANFROST_JD_REQ_FS (1 << 0) +#define PANFROST_JD_REQ_CYCLE_COUNT (1 << 1) /** * struct drm_panfrost_submit - ioctl argument for submitting commands to the 3D * engine.