diff mbox

IB/mthca: Fix how mthca_map_user_db() calls gup

Message ID 20180125192727.dkwuqel3xfut66wj@linux-n805 (mailing list archive)
State Accepted
Headers show

Commit Message

Davidlohr Bueso Jan. 25, 2018, 7:27 p.m. UTC
On Thu, 25 Jan 2018, Jason Gunthorpe wrote:

>>
>> Since the original post was referred to an ABBA deadlock, wouldn't we
>> have to drop db_tab->mutex, then grab both in the proper order?
>
>I had understood that was only a concern because Davidlohr was having
>trouble proving the callchain didn't include mmap_sem already..
>
>I can see the call chain all ends on verbs ops, and I know verbs ops
>with ucontext's are never called under mmap_sem by the core code..

Right. Ok so this simplifies things and we can just use gup_fast().

Thanks,
Davidlohr

---8<---------------------------------------------------
[PATCH v2] IB/mthca: Fix gup usage in mthca_map_user_db()

get_user_pages() must be called with mmap_sem held, currently
it is not. In fact it is called under the user db_table->mutex.
To fix this we can convert gup to use the fast alternative,
and safely avoid taking mmap_sem, if possible. Furthermore
this is safe wrt to the mutex as other callers that take the
lock (unmap and alloc_db) are not called under mmap_sem
(hence possible deadlock).

Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
---
 drivers/infiniband/hw/mthca/mthca_memfree.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Doug Ledford Jan. 26, 2018, 3:44 p.m. UTC | #1
On Thu, 2018-01-25 at 11:27 -0800, Davidlohr Bueso wrote:
> On Thu, 25 Jan 2018, Jason Gunthorpe wrote:
> 
> > > 
> > > Since the original post was referred to an ABBA deadlock, wouldn't we
> > > have to drop db_tab->mutex, then grab both in the proper order?
> > 
> > I had understood that was only a concern because Davidlohr was having
> > trouble proving the callchain didn't include mmap_sem already..
> > 
> > I can see the call chain all ends on verbs ops, and I know verbs ops
> > with ucontext's are never called under mmap_sem by the core code..
> 
> Right. Ok so this simplifies things and we can just use gup_fast().
> 
> Thanks,
> Davidlohr

Thanks, applied.
diff mbox

Patch

diff --git a/drivers/infiniband/hw/mthca/mthca_memfree.c b/drivers/infiniband/hw/mthca/mthca_memfree.c
index c6fe89d79248..9a412738d5c3 100644
--- a/drivers/infiniband/hw/mthca/mthca_memfree.c
+++ b/drivers/infiniband/hw/mthca/mthca_memfree.c
@@ -472,7 +472,7 @@  int mthca_map_user_db(struct mthca_dev *dev, struct mthca_uar *uar,
 		goto out;
 	}
 
-	ret = get_user_pages(uaddr & PAGE_MASK, 1, FOLL_WRITE, pages, NULL);
+	ret = get_user_pages_fast(uaddr & PAGE_MASK, 1, FOLL_WRITE, pages);
 	if (ret < 0)
 		goto out;