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=-14.0 required=3.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,URIBL_BLACK, URIBL_BLOCKED autolearn=unavailable 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 1AAF5C4361B for ; Tue, 15 Dec 2020 03:25:28 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id C998E223D6 for ; Tue, 15 Dec 2020 03:25:27 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1725785AbgLODZJ (ORCPT ); Mon, 14 Dec 2020 22:25:09 -0500 Received: from mail.kernel.org ([198.145.29.99]:36826 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727187AbgLODJq (ORCPT ); Mon, 14 Dec 2020 22:09:46 -0500 Date: Mon, 14 Dec 2020 19:08:30 -0800 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linux-foundation.org; s=korg; t=1608001712; bh=DnHKdkJcAQ7EZxFtmprjv6IzsJznAWdr1dgx+XQfzq0=; h=From:To:Subject:In-Reply-To:From; b=plGGw5wmbup08H+2R6B67wDOW0t84aQcXLR8BDhZqZ3UvTon0M6nlUtlAMDBDKV6z iuee99SJ7IeFAOPdI/CNJF7u+n98rI7vsZMCiV0NG0/MDLI5tSSp1JZ+bNV7SzCOnV GIbD5wCqNHIXO4UrD3tR9uA4SaAQqiN4Gw/dhaf0= From: Andrew Morton To: akpm@linux-foundation.org, bigeasy@linutronix.de, cai@lca.pw, christian.koenig@amd.com, cl@linux.com, daniel.vetter@ffwll.ch, daniel.vetter@intel.com, david@fromorbit.com, iamjoonsoo.kim@lge.com, jgg@mellanox.com, jgg@nvidia.com, linux-mm@kvack.org, longman@redhat.com, maarten.lankhorst@linux.intel.com, mathieu.desnoyers@efficios.com, mingo@kernel.org, mingo@redhat.com, mm-commits@vger.kernel.org, paulmck@kernel.org, penberg@kernel.org, peterz@infradead.org, rdunlap@infradead.org, rientjes@google.com, tglx@linutronix.de, thomas_os@shipmail.org, torvalds@linux-foundation.org, vbabka@suse.cz, walken@google.com, will@kernel.org, willy@infradead.org Subject: [patch 090/200] mm: track mmu notifiers in fs_reclaim_acquire/release Message-ID: <20201215030830.XBcE-svTH%akpm@linux-foundation.org> In-Reply-To: <20201214190237.a17b70ae14f129e2dca3d204@linux-foundation.org> User-Agent: s-nail v14.8.16 MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Precedence: bulk Reply-To: linux-kernel@vger.kernel.org List-ID: X-Mailing-List: mm-commits@vger.kernel.org =46rom: Daniel Vetter Subject: mm: track mmu notifiers in fs_reclaim_acquire/release fs_reclaim_acquire/release nicely catch recursion issues when allocating GFP_KERNEL memory against shrinkers (which gpu drivers tend to use to keep the excessive caches in check). For mmu notifier recursions we do have lockdep annotations since 23b68395c7c7 ("mm/mmu_notifiers: add a lockdep map for invalidate_range_start/end"). But these only fire if a path actually results in some pte invalidation - for most small allocations that's very rarely the case. The other trouble is that pte invalidation can happen any time when __GFP_RECLAIM is set.=20 Which means only really GFP_ATOMIC is a safe choice, GFP_NOIO isn't good enough to avoid potential mmu notifier recursion. I was pondering whether we should just do the general annotation, but there's always the risk for false positives. Plus I'm assuming that the core fs and io code is a lot better reviewed and tested than random mmu notifier code in drivers. Hence why I decide to only annotate for that specific case. Furthermore even if we'd create a lockdep map for direct reclaim, we'd still need to explicit pull in the mmu notifier map - there's a lot more places that do pte invalidation than just direct reclaim, these two contexts arent the same. Note that the mmu notifiers needing their own independent lockdep map is also the reason we can't hold them from fs_reclaim_acquire to fs_reclaim_release - it would nest with the acquistion in the pte invalidation code, causing a lockdep splat. And we can't remove the annotations from pte invalidation and all the other places since they're called from many other places than page reclaim. Hence we can only do the equivalent of might_lock, but on the raw lockdep map. With this we can also remove the lockdep priming added in 66204f1d2d1b ("mm/mmu_notifiers: prime lockdep") since the new annotations are strictly more powerful. Link: https://lkml.kernel.org/r/20201125162532.1299794-2-daniel.vetter@ffwl= l.ch Signed-off-by: Daniel Vetter Reviewed-by: Jason Gunthorpe Cc: Dave Chinner Cc: Qian Cai Cc: Thomas Hellstr=C3=B6m (Intel) Cc: Jason Gunthorpe Cc: Maarten Lankhorst Cc: Christian K=C3=B6nig Cc: "Matthew Wilcox (Oracle)" Cc: Christoph Lameter Cc: David Rientjes Cc: Ingo Molnar Cc: Ingo Molnar Cc: Joonsoo Kim Cc: Mathieu Desnoyers Cc: Michel Lespinasse Cc: Paul E. McKenney Cc: Pekka Enberg Cc: Peter Zijlstra (Intel) Cc: Randy Dunlap Cc: Sebastian Andrzej Siewior Cc: Thomas Gleixner Cc: Vlastimil Babka Cc: Waiman Long Cc: Will Deacon Signed-off-by: Andrew Morton --- mm/mmu_notifier.c | 7 ------- mm/page_alloc.c | 31 ++++++++++++++++++++----------- 2 files changed, 20 insertions(+), 18 deletions(-) --- a/mm/mmu_notifier.c~mm-track-mmu-notifiers-in-fs_reclaim_acquire-release +++ a/mm/mmu_notifier.c @@ -612,13 +612,6 @@ int __mmu_notifier_register(struct mmu_n mmap_assert_write_locked(mm); BUG_ON(atomic_read(&mm->mm_users) <=3D 0); =20 - if (IS_ENABLED(CONFIG_LOCKDEP)) { - fs_reclaim_acquire(GFP_KERNEL); - lock_map_acquire(&__mmu_notifier_invalidate_range_start_map); - lock_map_release(&__mmu_notifier_invalidate_range_start_map); - fs_reclaim_release(GFP_KERNEL); - } - if (!mm->notifier_subscriptions) { /* * kmalloc cannot be called under mm_take_all_locks(), but we --- a/mm/page_alloc.c~mm-track-mmu-notifiers-in-fs_reclaim_acquire-release +++ a/mm/page_alloc.c @@ -57,6 +57,7 @@ #include #include #include +#include #include #include #include @@ -4264,10 +4265,8 @@ should_compact_retry(struct alloc_contex static struct lockdep_map __fs_reclaim_map =3D STATIC_LOCKDEP_MAP_INIT("fs_reclaim", &__fs_reclaim_map); =20 -static bool __need_fs_reclaim(gfp_t gfp_mask) +static bool __need_reclaim(gfp_t gfp_mask) { - gfp_mask =3D current_gfp_context(gfp_mask); - /* no reclaim without waiting on it */ if (!(gfp_mask & __GFP_DIRECT_RECLAIM)) return false; @@ -4276,10 +4275,6 @@ static bool __need_fs_reclaim(gfp_t gfp_ if (current->flags & PF_MEMALLOC) return false; =20 - /* We're only interested __GFP_FS allocations for now */ - if (!(gfp_mask & __GFP_FS)) - return false; - if (gfp_mask & __GFP_NOLOCKDEP) return false; =20 @@ -4298,15 +4293,29 @@ void __fs_reclaim_release(void) =20 void fs_reclaim_acquire(gfp_t gfp_mask) { - if (__need_fs_reclaim(gfp_mask)) - __fs_reclaim_acquire(); + gfp_mask =3D current_gfp_context(gfp_mask); + + if (__need_reclaim(gfp_mask)) { + if (gfp_mask & __GFP_FS) + __fs_reclaim_acquire(); + +#ifdef CONFIG_MMU_NOTIFIER + lock_map_acquire(&__mmu_notifier_invalidate_range_start_map); + lock_map_release(&__mmu_notifier_invalidate_range_start_map); +#endif + + } } EXPORT_SYMBOL_GPL(fs_reclaim_acquire); =20 void fs_reclaim_release(gfp_t gfp_mask) { - if (__need_fs_reclaim(gfp_mask)) - __fs_reclaim_release(); + gfp_mask =3D current_gfp_context(gfp_mask); + + if (__need_reclaim(gfp_mask)) { + if (gfp_mask & __GFP_FS) + __fs_reclaim_release(); + } } EXPORT_SYMBOL_GPL(fs_reclaim_release); #endif _