[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20200520124817.GG31189@ziepe.ca>
Date: Wed, 20 May 2020 09:48:17 -0300
From: Jason Gunthorpe <jgg@...pe.ca>
To: Michel Lespinasse <walken@...gle.com>
Cc: John Hubbard <jhubbard@...dia.com>,
Matthew Wilcox <willy@...radead.org>,
Laurent Dufour <ldufour@...ux.ibm.com>,
Andrew Morton <akpm@...ux-foundation.org>,
linux-mm <linux-mm@...ck.org>,
LKML <linux-kernel@...r.kernel.org>,
Peter Zijlstra <peterz@...radead.org>,
Vlastimil Babka <vbabka@...e.cz>,
Liam Howlett <Liam.Howlett@...cle.com>,
Jerome Glisse <jglisse@...hat.com>,
Davidlohr Bueso <dave@...olabs.net>,
David Rientjes <rientjes@...gle.com>,
Hugh Dickins <hughd@...gle.com>, Ying Han <yinghan@...gle.com>,
Daniel Jordan <daniel.m.jordan@...cle.com>
Subject: Re: [PATCH v5.5 10/10] mmap locking API: rename mmap_sem to mmap_lock
On Tue, May 19, 2020 at 07:39:30PM -0700, Michel Lespinasse wrote:
> > > I think this assertion should be deleted from this driver. It's there
> > > in case get_user_pages_fast() takes the mmap sem. It would make sense to
> > > have this assertion in get_user_pages_fast() in case we take the fast path
> > > which doesn't acquire the mmap_sem. Something like this:
>
> I like this idea a lot - having might_lock assertions in
> get_user_pages_fast makes a log more sense than doing the same at the
> call sites.
+1 I've wanted to see more complete lockdep annotations in gup.c for a
while now.. There has been a number of bugs this would have caught
Jason
Powered by blists - more mailing lists