Received: by 2002:a6b:fb09:0:0:0:0:0 with SMTP id h9csp1664381iog; Sat, 25 Jun 2022 16:07:02 -0700 (PDT) X-Google-Smtp-Source: AGRyM1sjJm5SxTLHbwm5UOjnxC518K9p0o1TONhOAEEgRGrBvm8pPQaMjgyMlSOmObNXUl/ypR8n X-Received: by 2002:a63:1c5a:0:b0:3fd:1865:a789 with SMTP id c26-20020a631c5a000000b003fd1865a789mr5418715pgm.413.1656198421877; Sat, 25 Jun 2022 16:07:01 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1656198421; cv=none; d=google.com; s=arc-20160816; b=ggtWr5geT0T6YsfkhfIDPFL7uOVFUObSMydCBomZKiZYiqI9PxK0rVtYomW2xzsoQK dqUGtyTESQjbi6uDh6nFbD05mO2kPPHq7aeIx6NINpxk9Cs0DI8wclhDC0qxrulmzvem v3XXiRFQ+vaMrkWDWWyxRFBP+pG/Cho65hW+tQU8AcmWgLfH+EOccY+OYFujd9Di4usQ k6ENuD8lUvXYyn1rj4l5FnkrIRhlbSG6BX3mHNLZkBRyPpU6dqiXDjhyKrJqFqpSlSGH /7FYOPFDoLiWDL1AuNkpYaJBb26RPq3xXp9By7FfuAXoIEU78YLloQTMSs4SbXWLieX9 B3lQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:mime-version :references:in-reply-to:message-id:date:subject:cc:to:from :dkim-signature; bh=AeMYgqPNjsyamzteMz4JJUbia+3HcpDgXI4+3iGYjqA=; b=Jca6AMpuMfESdmNoNhLbxipXr/ICqxYBKS5qa+VZkBdRUy5qzrSmcWjSFERNxoqGdi f36hytV8czJZElQA3YnoejvJlR+xp9v3EY6uo7oBHqRifTzMSYnWDanoWWmFaLAreARF RlqAB0VKB+fgd7MEnf+u7ghD0MQvm/lUlGjosEJx3gTVxajR6fPBqxScpZ1ewpmIf/aM IqUQE7QILLI+0YMx1FKtBUC4Y2g4Qr6Wbo06nm/DI1OKzrurusGCgGZLBOuWzqCaKUnz Qzp8XvRUArQzS/PQC0RoqMMl5ZeEfmwiP2xpvDrPw9gDfrpIYUcRX+0UN7LA1d3tajof iMog== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@gmail.com header.s=20210112 header.b=XWwFhgsw; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=QUARANTINE dis=NONE) header.from=gmail.com Return-Path: Received: from out1.vger.email (out1.vger.email. [2620:137:e000::1:20]) by mx.google.com with ESMTP id b21-20020a170902d31500b0016a28448c46si7836039plc.561.2022.06.25.16.06.50; Sat, 25 Jun 2022 16:07:01 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20; Authentication-Results: mx.google.com; dkim=pass header.i=@gmail.com header.s=20210112 header.b=XWwFhgsw; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=QUARANTINE dis=NONE) header.from=gmail.com Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233681AbiFYWzq (ORCPT + 99 others); Sat, 25 Jun 2022 18:55:46 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:60692 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S233664AbiFYWz2 (ORCPT ); Sat, 25 Jun 2022 18:55:28 -0400 Received: from mail-pl1-x62a.google.com (mail-pl1-x62a.google.com [IPv6:2607:f8b0:4864:20::62a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id A6D4C13F44; Sat, 25 Jun 2022 15:55:27 -0700 (PDT) Received: by mail-pl1-x62a.google.com with SMTP id m14so5112989plg.5; Sat, 25 Jun 2022 15:55:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=AeMYgqPNjsyamzteMz4JJUbia+3HcpDgXI4+3iGYjqA=; b=XWwFhgsw0jy62ylwdGasi1YuRV0M1drEnIN3jNSmD9WAonjLsfbEA7hU4DteE1FAeW Ie+6qGqPtOARntKNsgRJstxzfWVM5MQyk4m8nh6cpAVhH2w7H2xiE7T/pzZpxCkUYcwu HEpM19RcB2tvIpKM1F8yvJqbid6mn8GN+sDQw/8i43aIqe28cYZzM8dSEvOYylmhX39/ t++myIfjITFWjSFJnrXu9VlSpUSHJ6JAYcpCxCSmLwwacVXTR4WUtVQF1ach9Fszoz02 tDq2kecCKTSUmI7ttzeMQoNIHRGkig8MqjZWHv6voqVmloRrqV3tyQ2VEHIMFDEfJ6+b ck1A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=AeMYgqPNjsyamzteMz4JJUbia+3HcpDgXI4+3iGYjqA=; b=33KnYIMh2Kq+U2XvPSiWUh/4l2HnJXHnDwmqtN3y2sttgvHWKSByHhJ4HM+6T38iZR kgZD2cX0P8lXKeMe+IWmZ1Ane2ib4KpWyWQ+QX/ULQ3uWz1LtUakv1UUFLg3vNlSiVCT SQ3FtXmAJCBUwX5pCp1H4m/BqpT+ktTw5vu8bldLqcC46EneHmGt+qL8YhW6silUecEl ROCDBIYUzrWxrlIpDLg5B+RSOPmcdy//2Prvfm8GqQlIZZlbvnwhp3VEdRD8Erp1riyd SU+iFY8+SO0u8fekAdonTQBEDZWqYL417BrnreduVcziyfABqOIezUewGPOzKeb2yRcq yfVg== X-Gm-Message-State: AJIora8sxrwMiRe5TnYec8gm1etnJqwr1bETO7qfg3kiXz3AP211ov1b bUJ1XQKVadYBjyzsBUmj0fE= X-Received: by 2002:a17:902:d2d1:b0:16a:1dd9:4d3d with SMTP id n17-20020a170902d2d100b0016a1dd94d3dmr6372390plc.18.1656197727296; Sat, 25 Jun 2022 15:55:27 -0700 (PDT) Received: from localhost ([2601:1c0:5200:a6:307:a401:7b76:c6e5]) by smtp.gmail.com with ESMTPSA id 9-20020a170902c20900b0015e8d4eb1dfsm4205062pll.41.2022.06.25.15.55.25 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 25 Jun 2022 15:55:25 -0700 (PDT) From: Rob Clark To: dri-devel@lists.freedesktop.org Cc: freedreno@lists.freedesktop.org, linux-arm-msm@vger.kernel.org, Rob Clark , Daniel Vetter , Thomas Zimmermann , Dmitry Osipenko , Maarten Lankhorst , Maxime Ripard , David Airlie , linux-kernel@vger.kernel.org (open list) Subject: [PATCH 11/15] drm/gem: Add LRU/shrinker helper Date: Sat, 25 Jun 2022 15:54:46 -0700 Message-Id: <20220625225454.81039-12-robdclark@gmail.com> X-Mailer: git-send-email 2.36.1 In-Reply-To: <20220625225454.81039-1-robdclark@gmail.com> References: <20220625225454.81039-1-robdclark@gmail.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Spam-Status: No, score=-2.1 required=5.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,FREEMAIL_FROM, RCVD_IN_DNSWL_NONE,SPF_HELO_NONE,SPF_PASS,T_SCC_BODY_TEXT_LINE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Rob Clark Add a simple LRU helper to assist with driver's shrinker implementation. It handles tracking the number of backing pages associated with a given LRU, and provides a helper to implement shrinker_scan. A driver can use multiple LRU instances to track objects in various states, for example a dontneed LRU for purgeable objects, a willneed LRU for evictable objects, and an unpinned LRU for objects without backing pages. All LRUs that the object can be moved between must share a single lock. Cc: Daniel Vetter Cc: Thomas Zimmermann Cc: Dmitry Osipenko Signed-off-by: Rob Clark --- drivers/gpu/drm/drm_gem.c | 183 ++++++++++++++++++++++++++++++++++++++ include/drm/drm_gem.h | 56 ++++++++++++ 2 files changed, 239 insertions(+) diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index eb0c2d041f13..684db28cc71c 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -165,6 +165,7 @@ void drm_gem_private_object_init(struct drm_device *dev, obj->resv = &obj->_resv; drm_vma_node_reset(&obj->vma_node); + INIT_LIST_HEAD(&obj->lru_node); } EXPORT_SYMBOL(drm_gem_private_object_init); @@ -951,6 +952,7 @@ drm_gem_object_release(struct drm_gem_object *obj) dma_resv_fini(&obj->_resv); drm_gem_free_mmap_offset(obj); + drm_gem_lru_remove(obj); } EXPORT_SYMBOL(drm_gem_object_release); @@ -1274,3 +1276,184 @@ drm_gem_unlock_reservations(struct drm_gem_object **objs, int count, ww_acquire_fini(acquire_ctx); } EXPORT_SYMBOL(drm_gem_unlock_reservations); + +/** + * drm_gem_lru_init - initialize a LRU + * + * @lru: The LRU to initialize + * @lock: The lock protecting the LRU + */ +void +drm_gem_lru_init(struct drm_gem_lru *lru, struct mutex *lock) +{ + lru->lock = lock; + lru->count = 0; + INIT_LIST_HEAD(&lru->list); +} +EXPORT_SYMBOL(drm_gem_lru_init); + +static void +lru_remove(struct drm_gem_object *obj) +{ + obj->lru->count -= obj->size >> PAGE_SHIFT; + WARN_ON(obj->lru->count < 0); + list_del(&obj->lru_node); + obj->lru = NULL; +} + +/** + * drm_gem_lru_remove - remove object from whatever LRU it is in + * + * If the object is currently in any LRU, remove it. + * + * @obj: The GEM object to remove from current LRU + */ +void +drm_gem_lru_remove(struct drm_gem_object *obj) +{ + struct drm_gem_lru *lru = obj->lru; + + if (!lru) + return; + + mutex_lock(lru->lock); + lru_remove(obj); + mutex_unlock(lru->lock); +} +EXPORT_SYMBOL(drm_gem_lru_remove); + +/** + * drm_gem_lru_move_tail - move the object to the tail of the LRU + * + * If the object is already in this LRU it will be moved to the + * tail. Otherwise it will be removed from whichever other LRU + * it is in (if any) and moved into this LRU. + * + * @lru: The LRU to move the object into. + * @obj: The GEM object to move into this LRU + */ +void +drm_gem_lru_move_tail(struct drm_gem_lru *lru, struct drm_gem_object *obj) +{ + mutex_lock(lru->lock); + drm_gem_lru_move_tail_locked(lru, obj); + mutex_unlock(lru->lock); +} +EXPORT_SYMBOL(drm_gem_lru_move_tail); + +/** + * drm_gem_lru_move_tail_locked - move the object to the tail of the LRU + * + * If the object is already in this LRU it will be moved to the + * tail. Otherwise it will be removed from whichever other LRU + * it is in (if any) and moved into this LRU. + * + * Call with LRU lock held. + * + * @lru: The LRU to move the object into. + * @obj: The GEM object to move into this LRU + */ +void +drm_gem_lru_move_tail_locked(struct drm_gem_lru *lru, struct drm_gem_object *obj) +{ + WARN_ON(!mutex_is_locked(lru->lock)); + + if (obj->lru) + lru_remove(obj); + + lru->count += obj->size >> PAGE_SHIFT; + list_add_tail(&obj->lru_node, &lru->list); + obj->lru = lru; +} +EXPORT_SYMBOL(drm_gem_lru_move_tail_locked); + +/** + * drm_gem_lru_scan - helper to implement shrinker.scan_objects + * + * If the shrink callback succeeds, it is expected that the driver + * move the object out of this LRU. + * + * If the LRU possibly contain active buffers, it is the responsibility + * of the shrink callback to check for this (ie. dma_resv_test_signaled()) + * or if necessary block until the buffer becomes idle. + * + * @lru: The LRU to scan + * @nr_to_scan: The number of pages to try to reclaim + * @shrink: Callback to try to shrink/reclaim the object. + */ +unsigned long +drm_gem_lru_scan(struct drm_gem_lru *lru, unsigned nr_to_scan, + bool (*shrink)(struct drm_gem_object *obj)) +{ + struct drm_gem_lru still_in_lru; + struct drm_gem_object *obj; + unsigned freed = 0; + + drm_gem_lru_init(&still_in_lru, lru->lock); + + mutex_lock(lru->lock); + + while (freed < nr_to_scan) { + obj = list_first_entry_or_null(&lru->list, typeof(*obj), lru_node); + + if (!obj) + break; + + drm_gem_lru_move_tail_locked(&still_in_lru, obj); + + /* + * If it's in the process of being freed, gem_object->free() + * may be blocked on lock waiting to remove it. So just + * skip it. + */ + if (!kref_get_unless_zero(&obj->refcount)) + continue; + + /* + * Now that we own a reference, we can drop the lock for the + * rest of the loop body, to reduce contention with other + * code paths that need the LRU lock + */ + mutex_unlock(lru->lock); + + /* + * Note that this still needs to be trylock, since we can + * hit shrinker in response to trying to get backing pages + * for this obj (ie. while it's lock is already held) + */ + if (!dma_resv_trylock(obj->resv)) + goto tail; + + if (shrink(obj)) { + freed += obj->size >> PAGE_SHIFT; + + /* + * If we succeeded in releasing the object's backing + * pages, we expect the driver to have moved the object + * out of this LRU + */ + WARN_ON(obj->lru == &still_in_lru); + WARN_ON(obj->lru == lru); + } + + dma_resv_unlock(obj->resv); + +tail: + drm_gem_object_put(obj); + mutex_lock(lru->lock); + } + + /* + * Move objects we've skipped over out of the temporary still_in_lru + * back into this LRU + */ + list_for_each_entry (obj, &still_in_lru.list, lru_node) + obj->lru = lru; + list_splice_tail(&still_in_lru.list, &lru->list); + lru->count += still_in_lru.count; + + mutex_unlock(lru->lock); + + return freed; +} +EXPORT_SYMBOL(drm_gem_lru_scan); diff --git a/include/drm/drm_gem.h b/include/drm/drm_gem.h index 87cffc9efa85..f13a9080af37 100644 --- a/include/drm/drm_gem.h +++ b/include/drm/drm_gem.h @@ -174,6 +174,41 @@ struct drm_gem_object_funcs { const struct vm_operations_struct *vm_ops; }; +/** + * struct drm_gem_lru - A simple LRU helper + * + * A helper for tracking GEM objects in a given state, to aid in + * driver's shrinker implementation. Tracks the count of pages + * for lockless &shrinker.count_objects, and provides + * &drm_gem_lru_scan for driver's &shrinker.scan_objects + * implementation. + */ +struct drm_gem_lru { + /** + * @lock: + * + * Lock protecting movement of GEM objects between LRUs. All + * LRUs that the object can move between should be protected + * by the same lock. + */ + struct mutex *lock; + + /** + * @count: + * + * The total number of backing pages of the GEM objects in + * this LRU. + */ + long count; + + /** + * @list: + * + * The LRU list. + */ + struct list_head list; +}; + /** * struct drm_gem_object - GEM buffer object * @@ -312,6 +347,20 @@ struct drm_gem_object { * */ const struct drm_gem_object_funcs *funcs; + + /** + * @lru_node: + * + * List node in a &drm_gem_lru. + */ + struct list_head lru_node; + + /** + * @lru: + * + * The current LRU list that the GEM object is on. + */ + struct drm_gem_lru *lru; }; /** @@ -420,4 +469,11 @@ void drm_gem_unlock_reservations(struct drm_gem_object **objs, int count, int drm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev, u32 handle, u64 *offset); +void drm_gem_lru_init(struct drm_gem_lru *lru, struct mutex *lock); +void drm_gem_lru_remove(struct drm_gem_object *obj); +void drm_gem_lru_move_tail(struct drm_gem_lru *lru, struct drm_gem_object *obj); +void drm_gem_lru_move_tail_locked(struct drm_gem_lru *lru, struct drm_gem_object *obj); +unsigned long drm_gem_lru_scan(struct drm_gem_lru *lru, unsigned nr_to_scan, + bool (*shrink)(struct drm_gem_object *obj)); + #endif /* __DRM_GEM_H__ */ -- 2.36.1