From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-alma10-1.taild15c8.ts.net [100.103.45.18]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 1BC973542CF; Mon, 3 Aug 2026 15:00:22 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=100.103.45.18 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1785769224; cv=none; b=km5U93zTa0p8V68c3IGMyotzjvgvKMFLz7un58kNfsw2NOuN7PzJxG05xV6Fy5APpDFYQ1fuLjyU4mWFCE1n/OAKCQLfEodB7QNjJGEoGmidE+F/++bY4q79CwIhWOAmIII/zg6CWCUp3ve6x7CaCVoP76fXlbO+kUAxC6VhRDw= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1785769224; c=relaxed/simple; bh=xn7HS/i4bXS7raN+YmRA7gDPOscAguRsatQPD6RRxRA=; h=Date:From:To:Cc:Subject:Message-ID:References:MIME-Version: Content-Type:Content-Disposition:In-Reply-To; b=BnEbjmU5zSEIPkDRg3lf31Lx2qMeLA/deDypQFmj7HXtiRNN/B1Xs+Q1hEXRP2F4q8dKLRp7N9dog1Qia8Wn0Vr7+lz4dME3OqqLgl6JBZ9KYDK2OArPJuMNuZP8AQNXij1XoVvU96ZG5rq9tSDmtR5+/hu9UW0hZV/1KUwonK8= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=Vik1jTt9; arc=none smtp.client-ip=100.103.45.18 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="Vik1jTt9" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 497791F000E9; Mon, 3 Aug 2026 15:00:19 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1785769222; bh=DYHyaHJJHP3pEUUb5RArEn5q/XsXxIfPC78T9whGzr4=; h=Date:From:To:Cc:Subject:References:In-Reply-To; b=Vik1jTt96uERvXxcpWMtBKyCMxM7oKJ8U3XxdDaLQjOA/eaiXi9ghWpojIC1nx5sj SInRjLBNlBxYVLqGczOlWQMaFdN7nqiY+qvXjmqmdz8kz7VZbEWVD9poNGm+i26c2M JK1AQu1n0NWlD/Cthka2vqxGKv+RqTy6RJCNnh4fGsjxYhe7JjkXDCFhonSQO/DjdM 7Wq3J1rNCZMF+d60g7RJOjJv9u1PMhlIym75jFvhPbqutLH5HW8cGvify3q8CTj5Y3 LSWGOaVywlWzPaS6TqltFmgDdeDHYBpn1Fz0VpobpJvslhtUb787FiEPCfDsP9oVun HxVT9AwBzssKg== Date: Mon, 3 Aug 2026 16:00:04 +0100 From: "Lorenzo Stoakes (ARM)" To: "Vlastimil Babka (SUSE)" Cc: Suren Baghdasaryan , akpm@linux-foundation.org, dave.hansen@linux.intel.com, Liam.Howlett@oracle.com, david@redhat.com, willy@infradead.org, shakeel.butt@linux.dev, 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 Subject: Re: [PATCH v3 3/5] mm: Add RCU-based VMA lookup helper that waits for writers Message-ID: References: <20260802215459.2769283-1-surenb@google.com> <20260802215459.2769283-4-surenb@google.com> <152ee467-b5e4-470c-ad95-c638aa9986bc@kernel.org> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <152ee467-b5e4-470c-ad95-c638aa9986bc@kernel.org> On Mon, Aug 03, 2026 at 04:55:19PM +0200, Vlastimil Babka (SUSE) wrote: > On 8/2/26 23:54, Suren Baghdasaryan wrote: > > From: Dave Hansen > > > > == Background == > > > > 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. > > > > == Problem == > > > > 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 = 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. > > > > == Solution == > > > > 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: > > Maybe mention that the helper is called vma_start_read_unlocked()? > > > > > 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 Hjønnevåg > > 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_struct *vma) > > } > > > > /* > > - * Use only while holding mmap read lock which guarantees that locking will not > > - * fail (nobody can concurrently write-lock the vma). vma_start_read() should > > + * 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 overflow. > > * This functionality is used to obtain vma read lock and drop the mmap read lock. > > + * VMA can't be detached while we are holding mmap lock, therefore in practice this > > + * function can fail only when there are so many readers that vm_refcnt overflows. > > */ > > 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(struct vm_area_struct *vma, int > > } > > > > /* > > - * Use only while holding mmap read lock which guarantees that locking will not > > - * fail (nobody can concurrently write-lock the vma). vma_start_read() should > > + * 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 overflow. > > * This functionality is used to obtain vma read lock and drop the mmap read lock. > > + * VMA can't be detached while we are holding mmap lock, therefore in practice this > > + * function can fail only when there are so many readers that vm_refcnt overflows. > > */ > > static inline bool vma_start_read_locked(struct vm_area_struct *vma) > > { > > return vma_start_read_locked_nested(vma, 0); > > } > > > > +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_struct *mm, > > return NULL; > > } > > > > +/* > > + * Find the VMA covering 'address' and lock it for reading. Waits for writers to > > + * finish if the VMA is being modified. Returns NULL if there is no VMA covering > > + * 'address'. > > Hm but it can also return NULL when vm_refcnt overflows, in theory. > Should we also return -EAGAIN (like uffd_lock_vma() below), or just retry in > here and hope for the best? The latter would be simpler for the users. > (AFAICS due to VM_REFCNT_LIMIT we never end up triggering the refcount > saturation) The problem is everything's unlocked so 'didn't find a VMA' doesn't really mean much more than 'something went wrong' because hey maybe if you check again now you'll find something :) So I think this might be a feature more than a bug, especially given overflow is not exactly likely. > > > + * > > + * 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 = 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 = vma_lookup(mm, address); > > + if (vma && !vma_start_read_locked(vma)) > > + vma = NULL; > > + mmap_read_unlock(mm); > > + > > + return vma; > > +} > > + > > static struct vm_area_struct *lock_next_vma_under_mmap_lock(struct mm_struct *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(struct mm_struct *mm, > > * > > * Should be called without holding mmap_lock. > > * > > - * Return: A locked vma containing @address, -ENOENT if no vma is found, or > > - * -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) > -- Cheers, Lorenzo