From patchwork Wed Sep 18 20:07:29 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Sean Paul X-Patchwork-Id: 11151211 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 60C9814ED for ; Wed, 18 Sep 2019 20:07:58 +0000 (UTC) Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 365EE21927 for ; Wed, 18 Sep 2019 20:07:58 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=lists.infradead.org header.i=@lists.infradead.org header.b="SEwQfh33"; dkim=fail reason="signature verification failed" (2048-bit key) header.d=poorly.run header.i=@poorly.run header.b="JQuBapSX" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 365EE21927 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=poorly.run Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-rockchip-bounces+patchwork-linux-rockchip=patchwork.kernel.org@lists.infradead.org DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20170209; h=Sender: Content-Transfer-Encoding:Content-Type:Cc:List-Subscribe:List-Help:List-Post: List-Archive:List-Unsubscribe:List-Id:MIME-Version:References:In-Reply-To: Message-Id:Date:Subject:To:From:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=Qk12j9TBwy4b9FhfF9KhHSnK10BESCPZpk1a969+jlM=; b=SEwQfh33JAiRTc kd8SIWt9eVrRHqANdWlkHd27/Pd14Hd1Nze175fmqRpgpUAqfZOxZ9j90MKPZPZx9Y/yZHPYYTbS5 viHDIfsk3MeDqnpvPS6O/cveqQa7E+XNfZ//eznNF1Mg2hCCezVqOorOZ8tLSeClje7L4C7CAWh6D EiaObOZB/i+A6p7+xvYZijXa/GtUl2d77N4q48jL0yzgo2sbfT0Yz2HD2HRqsTvUUh7UEXYAogsk0 NBFpXE/oCP519kVk0Pn2dJXy9k3bVBBCC2++uzumYyQRsFWWdrW8ryYg1bG9kECBrJOyIf7G9+RVu tGe9c7O1cX7FVungCtbA==; Received: from localhost ([127.0.0.1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.92.2 #3 (Red Hat Linux)) id 1iAgF0-0007zq-BC; Wed, 18 Sep 2019 20:07:54 +0000 Received: from mail-yb1-xb42.google.com ([2607:f8b0:4864:20::b42]) by bombadil.infradead.org with esmtps (Exim 4.92.2 #3 (Red Hat Linux)) id 1iAgEn-0007mO-CS for linux-rockchip@lists.infradead.org; Wed, 18 Sep 2019 20:07:43 +0000 Received: by mail-yb1-xb42.google.com with SMTP id y6so457167ybq.12 for ; Wed, 18 Sep 2019 13:07:40 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=poorly.run; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=8wRVg7Xp6X9fH0kFWOO4JmqVs8YGObxnFzNWoZ7A1I4=; b=JQuBapSX6N+q1mvLBlG+ko3vJHHAReQ7+RKU3yLX8PMtWnaVC4/ePwCDnv7GNJlyu7 5ufKPZlQCzxQ9sW8GoVcTdN27S1aYjJxUATQvTcg7SeRZX0lQiBzGTGnUvUp3OywcLL6 DwiqhbAREYbkAySZMFBtgEhdPeLwqOCPvoQwEKkMNomCPbtDc9pMmUvVAVmUAO+q4eM/ 1S8XNWIN8qsRv6pVHT+mU4v/LbO3hi/oO1q18koj2TDm4uLVJz2aaOOQFxwrlYQd663U RJa1Ph7YnozA8evZr52kZ9iUmxvSDyuRh7qlvLdxFop0HltioB8rvx/7+Uq2nOO90V7S O/8w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=8wRVg7Xp6X9fH0kFWOO4JmqVs8YGObxnFzNWoZ7A1I4=; b=OYSJIOkFmQ/LQ7cqtDYyCSPwSZBVbVhHBmxeqHkgfg3AYiZcKBjrquSpON3tuFjkeR CwonayBDvTNkq3mJWO5X0Ho36NA3plywulLLNxDc7wP8GcgtHyUZ+uLyF06xOhsKAg88 LjN+QF4oKiZbADyzuM7ZnGgSfViHG9HcUfRcwQBKcD6gAlK1R2GxmPfbAjo2tEF8WNuj YR+a0E2i7LgJLa4f/fItjc/LEavIaRkQ1axv2rkieRmTKJqg7yoVzrSaDmiPA1KR4ukT 4IOuRZXhT8hpfmLlsS7nj75DUFyYeduAxrh7JGErrMQ1M1xLdqHZlOWLd4GQ0g0GSDbU H3bA== X-Gm-Message-State: APjAAAXhDnTV84TsJokYVnR6jZcFl2Dncy9B90R3glolRc5bL+28kmm9 NnNFzBko/uScKFwzZgTwHJnpog== X-Google-Smtp-Source: APXvYqxp93kVw5JZ6ziN8xitJxe4JOB8xsrtmFJlgaaZHb9kWTb2WQcrBfBg/UL18FVpSe/EetmYyg== X-Received: by 2002:a25:2e44:: with SMTP id b4mr3154900ybn.9.1568837259968; Wed, 18 Sep 2019 13:07:39 -0700 (PDT) Received: from rosewood.cam.corp.google.com ([2620:0:1013:11:89c6:2139:5435:371d]) by smtp.gmail.com with ESMTPSA id 207sm1429317ywu.106.2019.09.18.13.07.39 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 18 Sep 2019 13:07:39 -0700 (PDT) From: Sean Paul To: dri-devel@lists.freedesktop.org Subject: [PATCH v2 2/3] drm: Measure Self Refresh Entry/Exit times to avoid thrashing Date: Wed, 18 Sep 2019 16:07:29 -0400 Message-Id: <20190918200734.149876-2-sean@poorly.run> X-Mailer: git-send-email 2.23.0.237.gc6a4ce50a0-goog In-Reply-To: <20190918200734.149876-1-sean@poorly.run> References: <20190918200734.149876-1-sean@poorly.run> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20190918_130741_534419_056FF669 X-CRM114-Status: GOOD ( 21.14 ) X-Spam-Score: -0.2 (/) X-Spam-Report: SpamAssassin version 3.4.2 on bombadil.infradead.org summary: Content analysis details: (-0.2 points) pts rule name description ---- ---------------------- -------------------------------------------------- -0.0 RCVD_IN_DNSWL_NONE RBL: Sender listed at https://www.dnswl.org/, no trust [2607:f8b0:4864:20:0:0:0:b42 listed in] [list.dnswl.org] 0.0 SPF_HELO_NONE SPF: HELO does not publish an SPF Record -0.0 SPF_PASS SPF: sender matches SPF record -0.1 DKIM_VALID Message has at least one valid DKIM or DK signature -0.1 DKIM_VALID_AU Message has a valid DKIM or DK signature from author's domain -0.1 DKIM_VALID_EF Message has a valid DKIM or DK signature from envelope-from domain 0.1 DKIM_SIGNED Message has a DKIM or DK signature, not necessarily valid X-BeenThere: linux-rockchip@lists.infradead.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Upstream kernel work for Rockchip platforms List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: heiko@sntech.de, linux-rockchip@lists.infradead.org, Maxime Ripard , Daniel Vetter , Maarten Lankhorst , Sandy Huang , linux-kernel@vger.kernel.org, David Airlie , Sean Paul , daniel@ffwll.ch, jekarl@iki.fi, Sean Paul , linux-arm-kernel@lists.infradead.org Sender: "Linux-rockchip" Errors-To: linux-rockchip-bounces+patchwork-linux-rockchip=patchwork.kernel.org@lists.infradead.org From: Sean Paul Currently the self refresh idle timer is a const set by the crtc. This is fine if the self refresh entry/exit times are well-known for all panels used on that crtc. However panels and workloads can vary quite a bit, and a timeout which works well for one doesn't work well for another. In the extreme, if the timeout is too short we could get in a situation where the self refresh exits are taking so long we queue up a self refresh entry before the exit commit is even finished. This patch changes the idle timeout to a moving average of the entry times + a moving average of exit times + the crtc constant. This patch was tested on rockchip, with a kevin CrOS panel the idle delay averages out to about ~235ms (35 entry + 100 exit + 100 const). On the same board, the bob panel idle delay lands around ~340ms (90 entry + 150 exit + 100 const). WRT the dedicated mutex in self_refresh_data, it would be nice if we could rely on drm_crtc.mutex to protect the average times, but there are a few reasons why a separate lock is a better choice: - We can't rely on drm_crtc.mutex being held if we're doing a nonblocking commit - We can't grab drm_crtc.mutex since drm_modeset_lock() doesn't tell us whether the lock was already held in the acquire context (it eats -EALREADY), so we can't tell if we should drop it or not - We don't need such a heavy-handed lock for what we're trying to do, commit ordering doesn't matter, so a point-of-use lock will be less contentious Reviewed-by: Daniel Vetter Signed-off-by: Sean Paul Link to v1: https://patchwork.freedesktop.org/patch/msgid/20190917200443.64481-2-sean@poorly.run Changes in v2: - Migrate locking explanation from comment to commit msg (Daniel) - Turf constant entry delay and multiply the avg times by 2 (Daniel) --- drivers/gpu/drm/drm_atomic_helper.c | 20 ++++++ drivers/gpu/drm/drm_self_refresh_helper.c | 72 +++++++++++++++++++-- drivers/gpu/drm/rockchip/rockchip_drm_vop.c | 5 +- include/drm/drm_self_refresh_helper.h | 6 +- 4 files changed, 90 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 9d7e4da6c292..3f13fa9a9e24 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -26,6 +26,7 @@ */ #include +#include #include #include @@ -1570,9 +1571,23 @@ static void commit_tail(struct drm_atomic_state *old_state) { struct drm_device *dev = old_state->dev; const struct drm_mode_config_helper_funcs *funcs; + ktime_t start; + s64 commit_time_ms; funcs = dev->mode_config.helper_private; + /* + * We're measuring the _entire_ commit, so the time will vary depending + * on how many fences and objects are involved. For the purposes of self + * refresh, this is desirable since it'll give us an idea of how + * congested things are. This will inform our decision on how often we + * should enter self refresh after idle. + * + * These times will be averaged out in the self refresh helpers to avoid + * overreacting over one outlier frame + */ + start = ktime_get(); + drm_atomic_helper_wait_for_fences(dev, old_state, false); drm_atomic_helper_wait_for_dependencies(old_state); @@ -1582,6 +1597,11 @@ static void commit_tail(struct drm_atomic_state *old_state) else drm_atomic_helper_commit_tail(old_state); + commit_time_ms = ktime_ms_delta(ktime_get(), start); + if (commit_time_ms > 0) + drm_self_refresh_helper_update_avg_times(old_state, + (unsigned long)commit_time_ms); + drm_atomic_helper_commit_cleanup_done(old_state); drm_atomic_state_put(old_state); diff --git a/drivers/gpu/drm/drm_self_refresh_helper.c b/drivers/gpu/drm/drm_self_refresh_helper.c index 9095cebf2147..68f4765a5896 100644 --- a/drivers/gpu/drm/drm_self_refresh_helper.c +++ b/drivers/gpu/drm/drm_self_refresh_helper.c @@ -5,6 +5,7 @@ * Authors: * Sean Paul */ +#include #include #include #include @@ -50,10 +51,17 @@ * atomic_check when &drm_crtc_state.self_refresh_active is true. */ +#define SELF_REFRESH_AVG_SEED_MS 200 + +DECLARE_EWMA(psr_time, 4, 4) + struct drm_self_refresh_data { struct drm_crtc *crtc; struct delayed_work entry_work; - unsigned int entry_delay_ms; + + struct mutex avg_mutex; + struct ewma_psr_time entry_avg_ms; + struct ewma_psr_time exit_avg_ms; }; static void drm_self_refresh_helper_entry_work(struct work_struct *work) @@ -121,6 +129,44 @@ static void drm_self_refresh_helper_entry_work(struct work_struct *work) drm_modeset_acquire_fini(&ctx); } +/** + * drm_self_refresh_helper_update_avg_times - Updates a crtc's SR time averages + * @state: the state which has just been applied to hardware + * @commit_time_ms: the amount of time in ms that this commit took to complete + * + * Called after &drm_mode_config_funcs.atomic_commit_tail, this function will + * update the average entry/exit self refresh times on self refresh transitions. + * These averages will be used when calculating how long to delay before + * entering self refresh mode after activity. + */ +void drm_self_refresh_helper_update_avg_times(struct drm_atomic_state *state, + unsigned int commit_time_ms) +{ + struct drm_crtc *crtc; + struct drm_crtc_state *old_crtc_state, *new_crtc_state; + int i; + + for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, + new_crtc_state, i) { + struct drm_self_refresh_data *sr_data = crtc->self_refresh_data; + struct ewma_psr_time *time; + + if (old_crtc_state->self_refresh_active == + new_crtc_state->self_refresh_active) + continue; + + if (new_crtc_state->self_refresh_active) + time = &sr_data->entry_avg_ms; + else + time = &sr_data->exit_avg_ms; + + mutex_lock(&sr_data->avg_mutex); + ewma_psr_time_add(time, commit_time_ms); + mutex_unlock(&sr_data->avg_mutex); + } +} +EXPORT_SYMBOL(drm_self_refresh_helper_update_avg_times); + /** * drm_self_refresh_helper_alter_state - Alters the atomic state for SR exit * @state: the state currently being checked @@ -152,6 +198,7 @@ void drm_self_refresh_helper_alter_state(struct drm_atomic_state *state) for_each_new_crtc_in_state(state, crtc, crtc_state, i) { struct drm_self_refresh_data *sr_data; + unsigned int delay; /* Don't trigger the entry timer when we're already in SR */ if (crtc_state->self_refresh_active) @@ -161,8 +208,13 @@ void drm_self_refresh_helper_alter_state(struct drm_atomic_state *state) if (!sr_data) continue; + mutex_lock(&sr_data->avg_mutex); + delay = (ewma_psr_time_read(&sr_data->entry_avg_ms) + + ewma_psr_time_read(&sr_data->exit_avg_ms)) * 2; + mutex_unlock(&sr_data->avg_mutex); + mod_delayed_work(system_wq, &sr_data->entry_work, - msecs_to_jiffies(sr_data->entry_delay_ms)); + msecs_to_jiffies(delay)); } } EXPORT_SYMBOL(drm_self_refresh_helper_alter_state); @@ -170,12 +222,10 @@ EXPORT_SYMBOL(drm_self_refresh_helper_alter_state); /** * drm_self_refresh_helper_init - Initializes self refresh helpers for a crtc * @crtc: the crtc which supports self refresh supported displays - * @entry_delay_ms: amount of inactivity to wait before entering self refresh * * Returns zero if successful or -errno on failure */ -int drm_self_refresh_helper_init(struct drm_crtc *crtc, - unsigned int entry_delay_ms) +int drm_self_refresh_helper_init(struct drm_crtc *crtc) { struct drm_self_refresh_data *sr_data = crtc->self_refresh_data; @@ -189,8 +239,18 @@ int drm_self_refresh_helper_init(struct drm_crtc *crtc, INIT_DELAYED_WORK(&sr_data->entry_work, drm_self_refresh_helper_entry_work); - sr_data->entry_delay_ms = entry_delay_ms; sr_data->crtc = crtc; + mutex_init(&sr_data->avg_mutex); + ewma_psr_time_init(&sr_data->entry_avg_ms); + ewma_psr_time_init(&sr_data->exit_avg_ms); + + /* + * Seed the averages so they're non-zero (and sufficiently large + * for even poorly performing panels). As time goes on, this will be + * averaged out and the values will trend to their true value. + */ + ewma_psr_time_add(&sr_data->entry_avg_ms, SELF_REFRESH_AVG_SEED_MS); + ewma_psr_time_add(&sr_data->exit_avg_ms, SELF_REFRESH_AVG_SEED_MS); crtc->self_refresh_data = sr_data; return 0; diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c index 2f821c58007c..613404f86668 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c @@ -39,8 +39,6 @@ #include "rockchip_drm_vop.h" #include "rockchip_rgb.h" -#define VOP_SELF_REFRESH_ENTRY_DELAY_MS 100 - #define VOP_WIN_SET(vop, win, name, v) \ vop_reg_set(vop, &win->phy->name, win->base, ~0, v, #name) #define VOP_SCL_SET(vop, win, name, v) \ @@ -1563,8 +1561,7 @@ static int vop_create_crtc(struct vop *vop) init_completion(&vop->line_flag_completion); crtc->port = port; - ret = drm_self_refresh_helper_init(crtc, - VOP_SELF_REFRESH_ENTRY_DELAY_MS); + ret = drm_self_refresh_helper_init(crtc); if (ret) DRM_DEV_DEBUG_KMS(vop->dev, "Failed to init %s with SR helpers %d, ignoring\n", diff --git a/include/drm/drm_self_refresh_helper.h b/include/drm/drm_self_refresh_helper.h index 397a583ccca7..5b79d253fb46 100644 --- a/include/drm/drm_self_refresh_helper.h +++ b/include/drm/drm_self_refresh_helper.h @@ -12,9 +12,9 @@ struct drm_atomic_state; struct drm_crtc; void drm_self_refresh_helper_alter_state(struct drm_atomic_state *state); +void drm_self_refresh_helper_update_avg_times(struct drm_atomic_state *state, + unsigned int commit_time_ms); -int drm_self_refresh_helper_init(struct drm_crtc *crtc, - unsigned int entry_delay_ms); - +int drm_self_refresh_helper_init(struct drm_crtc *crtc); void drm_self_refresh_helper_cleanup(struct drm_crtc *crtc); #endif