From: Ralph Campbell <rcampbell@nvidia.com>
To: Jason Gunthorpe <jgg@ziepe.ca>,
Jerome Glisse <jglisse@redhat.com>,
"John Hubbard" <jhubbard@nvidia.com>, <Felix.Kuehling@amd.com>
Cc: <linux-rdma@vger.kernel.org>, <linux-mm@kvack.org>,
Andrea Arcangeli <aarcange@redhat.com>,
<dri-devel@lists.freedesktop.org>,
<amd-gfx@lists.freedesktop.org>,
Jason Gunthorpe <jgg@mellanox.com>
Subject: Re: [PATCH v2 hmm 01/11] mm/hmm: fix use after free with struct hmm in the mmu notifiers
Date: Fri, 7 Jun 2019 11:12:14 -0700 [thread overview]
Message-ID: <377cadfa-180e-9a6a-49df-0c2c27ae6fb3@nvidia.com> (raw)
In-Reply-To: <20190606184438.31646-2-jgg@ziepe.ca>
On 6/6/19 11:44 AM, Jason Gunthorpe wrote:
> From: Jason Gunthorpe <jgg@mellanox.com>
>
> mmu_notifier_unregister_no_release() is not a fence and the mmu_notifier
> system will continue to reference hmm->mn until the srcu grace period
> expires.
>
> Resulting in use after free races like this:
>
> CPU0 CPU1
> __mmu_notifier_invalidate_range_start()
> srcu_read_lock
> hlist_for_each ()
> // mn == hmm->mn
> hmm_mirror_unregister()
> hmm_put()
> hmm_free()
> mmu_notifier_unregister_no_release()
> hlist_del_init_rcu(hmm-mn->list)
> mn->ops->invalidate_range_start(mn, range);
> mm_get_hmm()
> mm->hmm = NULL;
> kfree(hmm)
> mutex_lock(&hmm->lock);
>
> Use SRCU to kfree the hmm memory so that the notifiers can rely on hmm
> existing. Get the now-safe hmm struct through container_of and directly
> check kref_get_unless_zero to lock it against free.
>
> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
You can add
Reviewed-by: Ralph Campbell <rcampbell@nvidia.com>
> ---
> v2:
> - Spell 'free' properly (Jerome/Ralph)
> ---
> include/linux/hmm.h | 1 +
> mm/hmm.c | 25 +++++++++++++++++++------
> 2 files changed, 20 insertions(+), 6 deletions(-)
>
> diff --git a/include/linux/hmm.h b/include/linux/hmm.h
> index 092f0234bfe917..688c5ca7068795 100644
> --- a/include/linux/hmm.h
> +++ b/include/linux/hmm.h
> @@ -102,6 +102,7 @@ struct hmm {
> struct mmu_notifier mmu_notifier;
> struct rw_semaphore mirrors_sem;
> wait_queue_head_t wq;
> + struct rcu_head rcu;
> long notifiers;
> bool dead;
> };
> diff --git a/mm/hmm.c b/mm/hmm.c
> index 8e7403f081f44a..547002f56a163d 100644
> --- a/mm/hmm.c
> +++ b/mm/hmm.c
> @@ -113,6 +113,11 @@ static struct hmm *hmm_get_or_create(struct mm_struct *mm)
> return NULL;
> }
>
> +static void hmm_free_rcu(struct rcu_head *rcu)
> +{
> + kfree(container_of(rcu, struct hmm, rcu));
> +}
> +
> static void hmm_free(struct kref *kref)
> {
> struct hmm *hmm = container_of(kref, struct hmm, kref);
> @@ -125,7 +130,7 @@ static void hmm_free(struct kref *kref)
> mm->hmm = NULL;
> spin_unlock(&mm->page_table_lock);
>
> - kfree(hmm);
> + mmu_notifier_call_srcu(&hmm->rcu, hmm_free_rcu);
> }
>
> static inline void hmm_put(struct hmm *hmm)
> @@ -153,10 +158,14 @@ void hmm_mm_destroy(struct mm_struct *mm)
>
> static void hmm_release(struct mmu_notifier *mn, struct mm_struct *mm)
> {
> - struct hmm *hmm = mm_get_hmm(mm);
> + struct hmm *hmm = container_of(mn, struct hmm, mmu_notifier);
> struct hmm_mirror *mirror;
> struct hmm_range *range;
>
> + /* hmm is in progress to free */
> + if (!kref_get_unless_zero(&hmm->kref))
> + return;
> +
> /* Report this HMM as dying. */
> hmm->dead = true;
>
> @@ -194,13 +203,15 @@ static void hmm_release(struct mmu_notifier *mn, struct mm_struct *mm)
> static int hmm_invalidate_range_start(struct mmu_notifier *mn,
> const struct mmu_notifier_range *nrange)
> {
> - struct hmm *hmm = mm_get_hmm(nrange->mm);
> + struct hmm *hmm = container_of(mn, struct hmm, mmu_notifier);
> struct hmm_mirror *mirror;
> struct hmm_update update;
> struct hmm_range *range;
> int ret = 0;
>
> - VM_BUG_ON(!hmm);
> + /* hmm is in progress to free */
> + if (!kref_get_unless_zero(&hmm->kref))
> + return 0;
>
> update.start = nrange->start;
> update.end = nrange->end;
> @@ -245,9 +256,11 @@ static int hmm_invalidate_range_start(struct mmu_notifier *mn,
> static void hmm_invalidate_range_end(struct mmu_notifier *mn,
> const struct mmu_notifier_range *nrange)
> {
> - struct hmm *hmm = mm_get_hmm(nrange->mm);
> + struct hmm *hmm = container_of(mn, struct hmm, mmu_notifier);
>
> - VM_BUG_ON(!hmm);
> + /* hmm is in progress to free */
> + if (!kref_get_unless_zero(&hmm->kref))
> + return;
>
> mutex_lock(&hmm->lock);
> hmm->notifiers--;
>
next prev parent reply other threads:[~2019-06-07 18:12 UTC|newest]
Thread overview: 79+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-06-06 18:44 [PATCH v2 hmm 00/11] Various revisions from a locking/code review Jason Gunthorpe
2019-06-06 18:44 ` [PATCH v2 hmm 01/11] mm/hmm: fix use after free with struct hmm in the mmu notifiers Jason Gunthorpe
2019-06-07 2:29 ` John Hubbard
2019-06-07 12:34 ` Jason Gunthorpe
2019-06-07 13:42 ` Jason Gunthorpe
2019-06-08 1:13 ` John Hubbard
2019-06-08 1:37 ` John Hubbard
2019-06-07 18:12 ` Ralph Campbell [this message]
2019-06-08 8:49 ` Christoph Hellwig
2019-06-08 11:33 ` Jason Gunthorpe
2019-06-06 18:44 ` [PATCH v2 hmm 02/11] mm/hmm: Use hmm_mirror not mm as an argument for hmm_range_register Jason Gunthorpe
2019-06-07 2:36 ` John Hubbard
2019-06-07 18:24 ` Ralph Campbell
2019-06-07 22:39 ` Ralph Campbell
2019-06-10 13:09 ` Jason Gunthorpe
2019-06-07 22:33 ` Ira Weiny
2019-06-08 8:54 ` Christoph Hellwig
2019-06-11 19:44 ` Jason Gunthorpe
2019-06-12 7:12 ` Christoph Hellwig
2019-06-12 11:41 ` Jason Gunthorpe
2019-06-12 12:11 ` Christoph Hellwig
2019-06-06 18:44 ` [PATCH v2 hmm 03/11] mm/hmm: Hold a mmgrab from hmm to mm Jason Gunthorpe
2019-06-07 2:44 ` John Hubbard
2019-06-07 12:36 ` Jason Gunthorpe
2019-06-07 18:41 ` Ralph Campbell
2019-06-07 18:51 ` Jason Gunthorpe
2019-06-07 22:38 ` Ira Weiny
2019-06-06 18:44 ` [PATCH v2 hmm 04/11] mm/hmm: Simplify hmm_get_or_create and make it reliable Jason Gunthorpe
2019-06-07 2:54 ` John Hubbard
2019-06-07 18:52 ` Ralph Campbell
2019-06-07 22:44 ` Ira Weiny
2019-06-06 18:44 ` [PATCH v2 hmm 05/11] mm/hmm: Remove duplicate condition test before wait_event_timeout Jason Gunthorpe
2019-06-07 3:06 ` John Hubbard
2019-06-07 12:47 ` Jason Gunthorpe
2019-06-07 13:31 ` [PATCH v3 " Jason Gunthorpe
2019-06-07 22:55 ` Ira Weiny
2019-06-08 1:32 ` John Hubbard
2019-06-07 19:01 ` [PATCH v2 " Ralph Campbell
2019-06-07 19:13 ` Jason Gunthorpe
2019-06-07 20:21 ` Ralph Campbell
2019-06-07 20:44 ` Jason Gunthorpe
2019-06-07 22:13 ` Ralph Campbell
2019-06-08 1:47 ` Jason Gunthorpe
2019-06-06 18:44 ` [PATCH v2 hmm 06/11] mm/hmm: Hold on to the mmget for the lifetime of the range Jason Gunthorpe
2019-06-07 3:15 ` John Hubbard
2019-06-07 20:29 ` Ralph Campbell
2019-06-06 18:44 ` [PATCH v2 hmm 07/11] mm/hmm: Use lockdep instead of comments Jason Gunthorpe
2019-06-07 3:19 ` John Hubbard
2019-06-07 20:31 ` Ralph Campbell
2019-06-07 22:16 ` Souptick Joarder
2019-06-06 18:44 ` [PATCH v2 hmm 08/11] mm/hmm: Remove racy protection against double-unregistration Jason Gunthorpe
2019-06-07 3:29 ` John Hubbard
2019-06-07 13:57 ` Jason Gunthorpe
2019-06-07 20:33 ` Ralph Campbell
2019-06-06 18:44 ` [PATCH v2 hmm 09/11] mm/hmm: Poison hmm_range during unregister Jason Gunthorpe
2019-06-07 3:37 ` John Hubbard
2019-06-07 14:03 ` Jason Gunthorpe
2019-06-07 20:46 ` Ralph Campbell
2019-06-07 20:49 ` Jason Gunthorpe
2019-06-07 23:01 ` Ira Weiny
2019-06-06 18:44 ` [PATCH v2 hmm 10/11] mm/hmm: Do not use list*_rcu() for hmm->ranges Jason Gunthorpe
2019-06-07 3:40 ` John Hubbard
2019-06-07 20:49 ` Ralph Campbell
2019-06-07 22:11 ` Souptick Joarder
2019-06-07 23:02 ` Ira Weiny
2019-06-06 18:44 ` [PATCH v2 hmm 11/11] mm/hmm: Remove confusing comment and logic from hmm_release Jason Gunthorpe
2019-06-07 3:47 ` John Hubbard
2019-06-07 12:58 ` Jason Gunthorpe
2019-06-07 21:37 ` Ralph Campbell
2019-06-08 2:12 ` Jason Gunthorpe
2019-06-10 16:02 ` Jason Gunthorpe
2019-06-10 22:03 ` Ralph Campbell
2019-06-07 16:05 ` [PATCH v2 12/11] mm/hmm: Fix error flows in hmm_invalidate_range_start Jason Gunthorpe
2019-06-07 23:52 ` Ralph Campbell
2019-06-08 1:35 ` Jason Gunthorpe
2019-06-11 19:48 ` [PATCH v2 hmm 00/11] Various revisions from a locking/code review Jason Gunthorpe
2019-06-12 17:54 ` Kuehling, Felix
2019-06-12 21:49 ` Yang, Philip
2019-06-13 17:50 ` Jason Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=377cadfa-180e-9a6a-49df-0c2c27ae6fb3@nvidia.com \
--to=rcampbell@nvidia.com \
--cc=Felix.Kuehling@amd.com \
--cc=aarcange@redhat.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=jgg@mellanox.com \
--cc=jgg@ziepe.ca \
--cc=jglisse@redhat.com \
--cc=jhubbard@nvidia.com \
--cc=linux-mm@kvack.org \
--cc=linux-rdma@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox