From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pj1-f70.google.com (mail-pj1-f70.google.com [209.85.216.70]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 3DE1D3DCD86 for ; Sun, 2 Aug 2026 21:55:11 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.216.70 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1785707713; cv=none; b=U0IUPSaRgvSq51oZcykK8Rm6ZwvQgs5bHN++yCtonUEenCyX5Jp2fBrfCvGSLizp4Lbwy2A4BO6gY2MPfoInsDlAkRPT8VCgHzSgQUIZ7ev/SqaLlMY7Ut1tKyfCPXDfeVvkqzMttUztcic21EcjH71uhcUKhMO58vN2+FZws7s= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1785707713; c=relaxed/simple; bh=+YqZSzP/I2a03BPXHHL01Weg11cKi/MFACPBRAnB+Jk=; h=Date:In-Reply-To:Mime-Version:References:Message-ID:Subject:From: To:Cc:Content-Type; b=WO6Tv/xnn7W51uW7vKCb+kuoKjPRs+kV5wkcNaQqXg1rypb0ktblbEXmGLywJfUGDuDgRF84ypKVz1j8nUJBU7iAsETrYPhaFZ8Ie8OqFb44dioUsVVyRNrQL5nwAIXiGSezBo8YIPdZzMmci5BkdoDPOLk2mIf+UTfKd/oZziw= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com; spf=pass smtp.mailfrom=flex--surenb.bounces.google.com; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b=NRGRHJ4E; arc=none smtp.client-ip=209.85.216.70 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=google.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=flex--surenb.bounces.google.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=google.com header.i=@google.com header.b="NRGRHJ4E" Received: by mail-pj1-f70.google.com with SMTP id 98e67ed59e1d1-38ea32e57e2so1146999a91.1 for ; Sun, 02 Aug 2026 14:55:11 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20251104; t=1785707710; x=1786312510; darn=vger.kernel.org; h=content-transfer-encoding:content-type:cc:to:from:subject :message-id:references:mime-version:in-reply-to:date:from:to:cc :subject:date:message-id:reply-to:content-type; bh=PbBMXqyTQfOdEQP+TBH4NONBT8Q85IiQo7bFMy00X9U=; b=NRGRHJ4EaWt7f9EonWSdkutLRSqgmZXj9+ytdpE43RSjDD9RJrTbvZTG6/x954oumj 1kCWh5IQWd+bDnx785Y0jvUCiSrPUPFAs3rooWWFpUWHzCanC3LUoNUYsV+JRGWPL44g 046aPjmghgTUbwLqu1/VZWp/Ni696/u/u31+nt9rYRJk42arLBPCKZruddspAZYnXTaR VR2gQSl23fJPjH8VJ6/1xsdMmNUF+9ANPqpDPPZmMUQIO4qndVEU619ziVjwc+BSLEUS GvPlOSn7c/aKby7b5crJa4p7l0Uuq+dDOYD7s6WtZeTSQeEkNj6KHc1JCuq25dverjMo mSaQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1785707710; x=1786312510; h=content-transfer-encoding:content-type:cc:to:from:subject :message-id:references:mime-version:in-reply-to:date :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to :content-type; bh=PbBMXqyTQfOdEQP+TBH4NONBT8Q85IiQo7bFMy00X9U=; b=AdUgxVCRbo0a19UWIzcviWbaZPgam1KwAo4pJrxYn9l7aXANuBa/TCuWeayx8KK4Mv Aiq/2rcrr1zLQMLAnbZdlAi7q/SpXNCuPSDMgURnS2GBMUur6VhpRIGNT3ouhTCTSXV3 ZVVMLOvDHizaBoMlms7ilBjXmFGnsXvzrDnn3tHmZ0ZQ6h72bSWeYKXdKKY3MmHiu63e oSRO/Vb0GgwUyrCUv7JqQpWz9U1gyhimHtZ234x9z0wgLIXecaJDgMYzlCtT4RFJKVgX irSlG8MvPfoXOO39tv7sMS2dYKKNJBPKWKQy29CppyFiwf3sO1FkKxK6kTlpCs5kAlFU XrXg== X-Forwarded-Encrypted: i=1; AHgh+RpZ0v+bpebMeYCI1A9pE74mY07uMJ1JOh3695xScrqycR+X5JLtuAaOn2pGcIjpeZ/yUKgw3h01gwyI1jA=@vger.kernel.org X-Gm-Message-State: AOJu0YyvaneSE/EyBCn19v8zVycq0xZdRzjywU2ttm9YBjWQ7O9+hU2H mk7a4H8t05CCbIDpbz+EkbUx9ACzOH85ww/r7eS+XzGtnOeMsqGWALBMr6YwT8NWI12ewNT5F1Z fXnZHQg== X-Received: from dlbou23.prod.google.com ([2002:a05:7022:1117:b0:13e:6e68:2a1]) (user=surenb job=prod-delivery.src-stubby-dispatcher) by 2002:a17:90b:4d0c:b0:37f:c69d:ce69 with SMTP id 98e67ed59e1d1-38fbc432bb3mr6817011a91.10.1785707710250; Sun, 02 Aug 2026 14:55:10 -0700 (PDT) Date: Sun, 2 Aug 2026 14:54:57 -0700 In-Reply-To: <20260802215459.2769283-1-surenb@google.com> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: Mime-Version: 1.0 References: <20260802215459.2769283-1-surenb@google.com> X-Mailer: git-send-email 2.55.0.508.g3f0d502094-goog Message-ID: <20260802215459.2769283-4-surenb@google.com> Subject: [PATCH v3 3/5] mm: Add RCU-based VMA lookup helper that waits for writers From: Suren Baghdasaryan To: akpm@linux-foundation.org Cc: dave.hansen@linux.intel.com, Liam.Howlett@oracle.com, ljs@kernel.org, david@redhat.com, willy@infradead.org, shakeel.butt@linux.dev, vbabka@kernel.org, jannh@google.com, aliceryhl@google.com, arve@android.com, cmllamas@google.com, christian@brauner.io, tkjos@android.com, dsahern@kernel.org, davem@davemloft.net, gregkh@linuxfoundation.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, netdev@vger.kernel.org, surenb@google.com Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable From: Dave Hansen =3D=3D Background =3D=3D There are basically two parallel ways to look up a VMA: the traditional way, which is protected by mmap_read_lock, and the RCU-based per-VMA lock way which is based on RCU and refcounts. =3D=3D Problem =3D=3D The mmap_lock one is more straightforward to use but it has a big disadvantage in that it can not be mixed with page faults since those can take mmap_lock for read, which can deadlock when mixed with nested page faults and parallel writers. For example: mmap_read_lock(mm); // Another thread does mmap_write_lock(). // New mmap_lock readers are blocked. vma =3D vma_lookup(mm, address); // This deadlocks on mmap_read_lock() if it faults: copy_from_user(address); mmap_read_unlock(mm); The per-VMA lock can be mixed with faults, but they can fail and need to be able to fall back to the traditional way. =3D=3D Solution =3D=3D Add a variant of the RCU-based lookup that waits for writers. This is basically the same as the existing RCU-based lookup, but on a failure to lock it temporarily takes mmap_lock for read and waits for writers to finish before locking the VMA, dropping the mmap_lock and returning the locked VMA. This has some advantages: 1. Callers do not need to have a fallback path for when they collide with writers. 2. It can be used in contexts where page faults can happen because it can take the mmap_lock for read but never *holds* it. 3. Its fast path does not require taking mmap_lock for read. Basically, when applied correctly, this approach results in faster *and* simpler code. Signed-off-by: Dave Hansen Signed-off-by: Suren Baghdasaryan Cc: Suren Baghdasaryan Cc: Andrew Morton Cc: "Liam R. Howlett" Cc: Lorenzo Stoakes Cc: Vlastimil Babka Cc: Shakeel Butt Cc: linux-mm@kvack.org Cc: Greg Kroah-Hartman Cc: Arve Hj=C3=B8nnev=C3=A5g Cc: Todd Kjos Cc: Christian Brauner Cc: Carlos Llamas Cc: Alice Ryhl Cc: "David S. Miller" Cc: David Ahern Cc: netdev@vger.kernel.org --- include/linux/mmap_lock.h | 15 +++++++++++---- mm/mmap_lock.c | 29 +++++++++++++++++++++++++++++ mm/userfaultfd.c | 6 ++++-- 3 files changed, 44 insertions(+), 6 deletions(-) diff --git a/include/linux/mmap_lock.h b/include/linux/mmap_lock.h index eb32b482434e..fdd8f5cf5722 100644 --- a/include/linux/mmap_lock.h +++ b/include/linux/mmap_lock.h @@ -228,10 +228,12 @@ static inline void vma_refcount_put(struct vm_area_st= ruct *vma) } =20 /* - * Use only while holding mmap read lock which guarantees that locking wil= l not - * fail (nobody can concurrently write-lock the vma). vma_start_read() sho= uld + * Use only while holding mmap read lock which guarantees that vma lock is= not + * contended (nobody can concurrently write-lock the vma). vma_start_read(= ) should * not be used in such cases because it might fail due to mm_lock_seq over= flow. * This functionality is used to obtain vma read lock and drop the mmap re= ad lock. + * VMA can't be detached while we are holding mmap lock, therefore in prac= tice this + * function can fail only when there are so many readers that vm_refcnt ov= erflows. */ static inline bool vma_start_read_locked_nested(struct vm_area_struct *vma= , int subclass) { @@ -247,16 +249,21 @@ static inline bool vma_start_read_locked_nested(struc= t vm_area_struct *vma, int } =20 /* - * Use only while holding mmap read lock which guarantees that locking wil= l not - * fail (nobody can concurrently write-lock the vma). vma_start_read() sho= uld + * Use only while holding mmap read lock which guarantees that vma lock is= not + * contended (nobody can concurrently write-lock the vma). vma_start_read(= ) should * not be used in such cases because it might fail due to mm_lock_seq over= flow. * This functionality is used to obtain vma read lock and drop the mmap re= ad lock. + * VMA can't be detached while we are holding mmap lock, therefore in prac= tice this + * function can fail only when there are so many readers that vm_refcnt ov= erflows. */ static inline bool vma_start_read_locked(struct vm_area_struct *vma) { return vma_start_read_locked_nested(vma, 0); } =20 +struct vm_area_struct *vma_start_read_unlocked(struct mm_struct *mm, + unsigned long address); + static inline void vma_end_read(struct vm_area_struct *vma) { vma_refcount_put(vma); diff --git a/mm/mmap_lock.c b/mm/mmap_lock.c index e20d01e8d38f..6ff05e68e61b 100644 --- a/mm/mmap_lock.c +++ b/mm/mmap_lock.c @@ -338,6 +338,35 @@ struct vm_area_struct *lock_vma_under_rcu(struct mm_st= ruct *mm, return NULL; } =20 +/* + * Find the VMA covering 'address' and lock it for reading. Waits for writ= ers to + * finish if the VMA is being modified. Returns NULL if there is no VMA co= vering + * 'address'. + * + * Use only in code paths where no mmap_lock and no VMA lock is held. + * + * The fast path does not take mmap_lock. + */ +struct vm_area_struct *vma_start_read_unlocked(struct mm_struct *mm, + unsigned long address) +{ + struct vm_area_struct *vma; + + /* Fast path: return stable VMA covering 'address': */ + vma =3D lock_vma_under_rcu(mm, address); + if (vma) + return vma; + + /* Slow path: preclude VMA writers by temporarily getting mmap read lock.= */ + mmap_read_lock(mm); + vma =3D vma_lookup(mm, address); + if (vma && !vma_start_read_locked(vma)) + vma =3D NULL; + mmap_read_unlock(mm); + + return vma; +} + static struct vm_area_struct *lock_next_vma_under_mmap_lock(struct mm_stru= ct *mm, struct vma_iterator *vmi, unsigned long from_addr) diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index edd90892f8cc..c3a0c38a3dc3 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -129,8 +129,10 @@ struct vm_area_struct *find_vma_and_prepare_anon(struc= t mm_struct *mm, * * Should be called without holding mmap_lock. * - * Return: A locked vma containing @address, -ENOENT if no vma is found, o= r - * -ENOMEM if anon_vma couldn't be allocated. + * Return: A locked vma containing @address, -ENOENT if no vma is found, + * -ENOMEM if anon_vma couldn't be allocated, or -EAGAIN if vma refcount + * overflow happened due to high number of readers and the caller should + * retry later. */ static struct vm_area_struct *uffd_lock_vma(struct mm_struct *mm, unsigned long address) --=20 2.55.0.508.g3f0d502094-goog