From patchwork Thu Aug 22 14:50:27 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Vetter X-Patchwork-Id: 11109519 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 4DBE4112C for ; Thu, 22 Aug 2019 14:50:40 +0000 (UTC) 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 mail.kernel.org (Postfix) with ESMTPS id 34E83233A0 for ; Thu, 22 Aug 2019 14:50:40 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 34E83233A0 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=ffwll.ch Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=intel-gfx-bounces@lists.freedesktop.org Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 5300E6E4F8; Thu, 22 Aug 2019 14:50:39 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mail-ed1-x544.google.com (mail-ed1-x544.google.com [IPv6:2a00:1450:4864:20::544]) by gabe.freedesktop.org (Postfix) with ESMTPS id 353F76E4F8 for ; Thu, 22 Aug 2019 14:50:37 +0000 (UTC) Received: by mail-ed1-x544.google.com with SMTP id z51so8296939edz.13 for ; Thu, 22 Aug 2019 07:50:37 -0700 (PDT) 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=mNvarGM4cBUzSRVeydoiHDGdjfOAF/fmkaXUCRxtFdk=; b=bAVs47osnXX090Ocb4ZrCRTmygk+K2s5QA6TPIC3N10bLagbjYwVJo5Hd5EEc2apDY twzB/YSW6AciOSQVankWAwdXcQkvli72i+K7ZBBh0EEi8HLcx1bDJJO4C80VXY0l3pb7 pD8YHthh8WRsO5AeuYHokm/EzPtmQPeo+0t+sLmYJgmZ8AX0JW5J7qQ2vxrkZPEEdS9N cl44M2W/hJegISjbhAvXdVT8N7yDfdqd/1fIY6ff8JoX1kylG9HGSbvnlZKq4HDpRGoS DP0vy1uXVNX5rn+3QfdG+QkcylutKm+TFR+cbvW6ZLhLgkNf51MDKmd80P+OQXBa+hRY 9zrg== X-Gm-Message-State: APjAAAW5e9GI2Pe86V8R044TvZL+5TuDNpL5xLjr9VFCRjSgOD9Ulo/I +wnikCzqHAkGbwWycbTLXNa7rRjpwfequw== X-Google-Smtp-Source: APXvYqydQ02RTgYdadFK6YhUVfKO9GxV8m4Jg70KrEDRicetcpuH9MHGhoaozAgvvsqXG2pNyLDIug== X-Received: by 2002:a17:906:f211:: with SMTP id gt17mr35455358ejb.263.1566485435251; Thu, 22 Aug 2019 07:50:35 -0700 (PDT) Received: from phenom.ffwll.local (212-51-149-96.fiber7.init7.net. [212.51.149.96]) by smtp.gmail.com with ESMTPSA id b18sm3689726eju.0.2019.08.22.07.50.34 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 22 Aug 2019 07:50:34 -0700 (PDT) From: Daniel Vetter To: Intel Graphics Development Date: Thu, 22 Aug 2019 16:50:27 +0200 Message-Id: <20190822145027.12572-1-daniel.vetter@ffwll.ch> X-Mailer: git-send-email 2.23.0.rc1 In-Reply-To: <20190820081951.25053-1-daniel.vetter@ffwll.ch> References: <20190820081951.25053-1-daniel.vetter@ffwll.ch> MIME-Version: 1.0 X-Mailman-Original-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ffwll.ch; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=mNvarGM4cBUzSRVeydoiHDGdjfOAF/fmkaXUCRxtFdk=; b=fv/iCbY6RM9IvU1NuBLS5qgvIkdL4iO2T6IR6bEfCyqv4A7meq6czfioqP6TDJ5YfA Cs6RpwUdvCilqBQ1bDXzJrBouRTdj0JjRVvIgz5sZKh8KJ7wyIWC0cncQ6ULwTnCxnnB hCl7DGI0fWfRMGsT2129tsKVany2ZnZ9ed/yk= Subject: [Intel-gfx] [PATCH] drm/i915: Switch obj->mm.lock lockdep annotations on its head X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Daniel Vetter , Daniel Vetter Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" The trouble with having a plain nesting flag for locks which do not naturally nest (unlike block devices and their partitions, which is the original motivation for nesting levels) is that lockdep will never spot a true deadlock if you screw up. This patch is an attempt at trying better, by highlighting a bit more the actual nature of the nesting that's going on. Essentially we have two kinds of objects: - objects without pages allocated, which cannot be on any lru and are hence inaccessible to the shrinker. - objects which have pages allocated, which are on an lru, and which the shrinker can decide to throw out. For the former type of object, memory allcoations while holding obj->mm.lock are permissible. For the latter they are not. And get/put_pages transitions between the two types of objects. This is still not entirely fool-proof since the rules might chance. But as long as we run such a code ever at runtime lockdep should be able to observe the inconsistency and complain (like with any other lockdep class that we've split up in multiple classes). But there are a few clear benefits: - We can drop the nesting flag parameter from __i915_gem_object_put_pages, because that function by definition is never going allocate memory, and calling it on an object which doesn't have its pages allocated would be a bug. - We strictly catch more bugs, since there's not only one place in the entire tree which is annotated with the special class. All the other places that had explicit lockdep nesting annotations we're now going to leave up to lockdep again. - Specifically this catches stuff like calling get_pages from put_pages (which isn't really a good idea, if we can call put_pages so could the shrinker). I've seen patches do exactly that. Of course I fully expect CI will show me for the fool I am with this one here :-) v2: There can only be one (lockdep only has a cache for the first subclass, not for deeper ones, and we don't want to make these locks even slower). Still separate enums for better documentation. Real fix: don forget about phys objs and pin_map(), and fix the shrinker to have the right annotations ... silly me. v3: Forgot usertptr too ... v4: Improve comment for pages_pin_count, drop the IMPORTANT comment and instead prime lockdep (Chris). v5: Appease checkpatch, no double empty lines (Chris) Cc: Chris Wilson Cc: "Tang, CQ" Cc: Tvrtko Ursulin Cc: Joonas Lahtinen Reviewed-by: Chris Wilson Signed-off-by: Daniel Vetter --- drivers/gpu/drm/i915/gem/i915_gem_object.c | 12 +++++++++++- drivers/gpu/drm/i915/gem/i915_gem_object.h | 16 +++++++++++++--- drivers/gpu/drm/i915/gem/i915_gem_object_types.h | 6 +++++- drivers/gpu/drm/i915/gem/i915_gem_pages.c | 9 ++++----- drivers/gpu/drm/i915/gem/i915_gem_phys.c | 2 +- drivers/gpu/drm/i915/gem/i915_gem_shrinker.c | 5 ++--- drivers/gpu/drm/i915/gem/i915_gem_userptr.c | 4 ++-- drivers/gpu/drm/i915/gem/selftests/huge_pages.c | 12 ++++++------ 8 files changed, 44 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index d7855dc5a5c5..1bdd7485bc72 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -22,6 +22,8 @@ * */ +#include + #include "display/intel_frontbuffer.h" #include "gt/intel_gt.h" #include "i915_drv.h" @@ -51,6 +53,14 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj, { mutex_init(&obj->mm.lock); + if (IS_ENABLED(CONFIG_LOCKDEP)) { + mutex_lock_nested(&obj->mm.lock, I915_MM_GET_PAGES); + fs_reclaim_acquire(GFP_KERNEL); + might_lock(&obj->mm.lock); + fs_reclaim_release(GFP_KERNEL); + mutex_unlock(&obj->mm.lock); + } + spin_lock_init(&obj->vma.lock); INIT_LIST_HEAD(&obj->vma.list); @@ -176,7 +186,7 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915, GEM_BUG_ON(!list_empty(&obj->lut_list)); atomic_set(&obj->mm.pages_pin_count, 0); - __i915_gem_object_put_pages(obj, I915_MM_NORMAL); + __i915_gem_object_put_pages(obj); GEM_BUG_ON(i915_gem_object_has_pages(obj)); bitmap_free(obj->bit_17); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index 5efb9936e05b..a0b1fa8a3224 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -281,11 +281,21 @@ i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj) enum i915_mm_subclass { /* lockdep subclass for obj->mm.lock/struct_mutex */ I915_MM_NORMAL = 0, - I915_MM_SHRINKER /* called "recursively" from direct-reclaim-esque */ + /* + * Only used by struct_mutex, when called "recursively" from + * direct-reclaim-esque. Safe because there is only every one + * struct_mutex in the entire system. */ + I915_MM_SHRINKER = 1, + /* + * Used for obj->mm.lock when allocating pages. Safe because the object + * isn't yet on any LRU, and therefore the shrinker can't deadlock on + * it. As soon as the object has pages, obj->mm.lock nests within + * fs_reclaim. + */ + I915_MM_GET_PAGES = 1, }; -int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj, - enum i915_mm_subclass subclass); +int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj); void i915_gem_object_truncate(struct drm_i915_gem_object *obj); void i915_gem_object_writeback(struct drm_i915_gem_object *obj); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h index ede0eb4218a8..7b7cf711a21a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object_types.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object_types.h @@ -156,7 +156,11 @@ struct drm_i915_gem_object { unsigned int pin_global; struct { - struct mutex lock; /* protects the pages and their use */ + /* + * Protects the pages and their use. Do not use directly, but + * instead go through the pin/unpin interfaces. + */ + struct mutex lock; atomic_t pages_pin_count; struct sg_table *pages; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c index 18f0ce0135c1..202526e8910f 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c @@ -101,7 +101,7 @@ int __i915_gem_object_get_pages(struct drm_i915_gem_object *obj) { int err; - err = mutex_lock_interruptible(&obj->mm.lock); + err = mutex_lock_interruptible_nested(&obj->mm.lock, I915_MM_GET_PAGES); if (err) return err; @@ -179,8 +179,7 @@ __i915_gem_object_unset_pages(struct drm_i915_gem_object *obj) return pages; } -int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj, - enum i915_mm_subclass subclass) +int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj) { struct sg_table *pages; int err; @@ -191,7 +190,7 @@ int __i915_gem_object_put_pages(struct drm_i915_gem_object *obj, GEM_BUG_ON(atomic_read(&obj->bind_count)); /* May be called by shrinker from within get_pages() (on another bo) */ - mutex_lock_nested(&obj->mm.lock, subclass); + mutex_lock(&obj->mm.lock); if (unlikely(atomic_read(&obj->mm.pages_pin_count))) { err = -EBUSY; goto unlock; @@ -285,7 +284,7 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj, if (unlikely(!i915_gem_object_has_struct_page(obj))) return ERR_PTR(-ENXIO); - err = mutex_lock_interruptible(&obj->mm.lock); + err = mutex_lock_interruptible_nested(&obj->mm.lock, I915_MM_GET_PAGES); if (err) return ERR_PTR(err); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/i915_gem_phys.c index 768356908160..2aea8960f0f1 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_phys.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_phys.c @@ -163,7 +163,7 @@ int i915_gem_object_attach_phys(struct drm_i915_gem_object *obj, int align) if (err) return err; - mutex_lock(&obj->mm.lock); + mutex_lock_nested(&obj->mm.lock, I915_MM_GET_PAGES); if (obj->mm.madv != I915_MADV_WILLNEED) { err = -EFAULT; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c index edd21d14e64f..0b0d6e27b996 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c @@ -98,7 +98,7 @@ static bool unsafe_drop_pages(struct drm_i915_gem_object *obj, flags = I915_GEM_OBJECT_UNBIND_ACTIVE; if (i915_gem_object_unbind(obj, flags) == 0) - __i915_gem_object_put_pages(obj, I915_MM_SHRINKER); + __i915_gem_object_put_pages(obj); return !i915_gem_object_has_pages(obj); } @@ -254,8 +254,7 @@ i915_gem_shrink(struct drm_i915_private *i915, if (unsafe_drop_pages(obj, shrink)) { /* May arrive from get_pages on another bo */ - mutex_lock_nested(&obj->mm.lock, - I915_MM_SHRINKER); + mutex_lock(&obj->mm.lock); if (!i915_gem_object_has_pages(obj)) { try_to_writeback(obj, shrink); count += obj->base.size >> PAGE_SHIFT; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c index 70dc506a5426..f3b3bc7c32cb 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c @@ -158,7 +158,7 @@ userptr_mn_invalidate_range_start(struct mmu_notifier *_mn, ret = i915_gem_object_unbind(obj, I915_GEM_OBJECT_UNBIND_ACTIVE); if (ret == 0) - ret = __i915_gem_object_put_pages(obj, I915_MM_SHRINKER); + ret = __i915_gem_object_put_pages(obj); i915_gem_object_put(obj); if (ret) goto unlock; @@ -514,7 +514,7 @@ __i915_gem_userptr_get_pages_worker(struct work_struct *_work) } } - mutex_lock(&obj->mm.lock); + mutex_lock_nested(&obj->mm.lock, I915_MM_GET_PAGES); if (obj->userptr.work == &work->work) { struct sg_table *pages = ERR_PTR(ret); diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c index 8de83c6d81f5..81af85971856 100644 --- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c +++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c @@ -562,7 +562,7 @@ static int igt_mock_ppgtt_misaligned_dma(void *arg) i915_vma_close(vma); i915_gem_object_unpin_pages(obj); - __i915_gem_object_put_pages(obj, I915_MM_NORMAL); + __i915_gem_object_put_pages(obj); i915_gem_object_put(obj); } @@ -590,7 +590,7 @@ static void close_object_list(struct list_head *objects, list_del(&obj->st_link); i915_gem_object_unpin_pages(obj); - __i915_gem_object_put_pages(obj, I915_MM_NORMAL); + __i915_gem_object_put_pages(obj); i915_gem_object_put(obj); } } @@ -860,7 +860,7 @@ static int igt_mock_ppgtt_64K(void *arg) i915_vma_close(vma); i915_gem_object_unpin_pages(obj); - __i915_gem_object_put_pages(obj, I915_MM_NORMAL); + __i915_gem_object_put_pages(obj); i915_gem_object_put(obj); } } @@ -1164,7 +1164,7 @@ static int igt_ppgtt_exhaust_huge(void *arg) } i915_gem_object_unpin_pages(obj); - __i915_gem_object_put_pages(obj, I915_MM_NORMAL); + __i915_gem_object_put_pages(obj); i915_gem_object_put(obj); } } @@ -1226,7 +1226,7 @@ static int igt_ppgtt_internal_huge(void *arg) } i915_gem_object_unpin_pages(obj); - __i915_gem_object_put_pages(obj, I915_MM_NORMAL); + __i915_gem_object_put_pages(obj); i915_gem_object_put(obj); } @@ -1295,7 +1295,7 @@ static int igt_ppgtt_gemfs_huge(void *arg) } i915_gem_object_unpin_pages(obj); - __i915_gem_object_put_pages(obj, I915_MM_NORMAL); + __i915_gem_object_put_pages(obj); i915_gem_object_put(obj); }