From: Jason Gunthorpe <jgg@nvidia.com>
To: Leon Romanovsky <leon@kernel.org>
Cc: Doug Ledford <dledford@redhat.com>, Shay Drory <shayd@nvidia.com>,
linux-rdma@vger.kernel.org
Subject: Re: [PATCH rdma-next] RDMA/restrack: Delay QP deletion till all users are gone
Date: Tue, 20 Apr 2021 12:25:41 -0300 [thread overview]
Message-ID: <20210420152541.GC1370958@nvidia.com> (raw)
In-Reply-To: <YH7Ru5ZSr1kWGZoa@unreal>
On Tue, Apr 20, 2021 at 04:06:03PM +0300, Leon Romanovsky wrote:
> On Tue, Apr 20, 2021 at 09:39:50AM -0300, Jason Gunthorpe wrote:
> > On Sun, Apr 18, 2021 at 04:37:35PM +0300, Leon Romanovsky wrote:
> > > From: Shay Drory <shayd@nvidia.com>
> > >
> > > Currently, in case of QP, the following use-after-free is possible:
> > >
> > > cpu0 cpu1
> > > res_get_common_dumpit()
> > > rdma_restrack_get()
> > > fill_res_qp_entry()
> > > ib_destroy_qp_user()
> > > rdma_restrack_del()
> > > qp->device->ops.destroy_qp()
> > > ib_query_qp()
> > > qp->device->ops.query_qp()
> > >
> > > This is because rdma_restrack_del(), in case of QP, isn't waiting until
> > > all users are gone.
> > >
> > > Fix it by making rdma_restrack_del() wait until all users are gone for
> > > QPs as well.
> > >
> > > Fixes: 13ef5539def7 ("RDMA/restrack: Count references to the verbs objects")
> > > Signed-off-by: Shay Drory <shayd@nvidia.com>
> > > Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
> > > drivers/infiniband/core/restrack.c | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/infiniband/core/restrack.c b/drivers/infiniband/core/restrack.c
> > > index ffabaf327242..def0c5b0efe9 100644
> > > +++ b/drivers/infiniband/core/restrack.c
> > > @@ -340,7 +340,7 @@ void rdma_restrack_del(struct rdma_restrack_entry *res)
> > > rt = &dev->res[res->type];
> > >
> > > old = xa_erase(&rt->xa, res->id);
> > > - if (res->type == RDMA_RESTRACK_MR || res->type == RDMA_RESTRACK_QP)
> > > + if (res->type == RDMA_RESTRACK_MR)
> > > return;
> >
> > Why is MR skipping this?
>
> I don't remember the justification for RDMA_RESTRACK_MR || RDMA_RESTRACK_QP check.
> My guess that it is related to the allocation flow (both are not converted) and
> have internal objects to the drivers.
Well, I don't understand why QP has a bug but MR would not have the
same one??
Jason
next prev parent reply other threads:[~2021-04-20 15:25 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-18 13:37 [PATCH rdma-next] RDMA/restrack: Delay QP deletion till all users are gone Leon Romanovsky
2021-04-20 12:39 ` Jason Gunthorpe
2021-04-20 13:06 ` Leon Romanovsky
2021-04-20 15:25 ` Jason Gunthorpe [this message]
2021-04-21 5:03 ` Leon Romanovsky
2021-04-22 14:29 ` Jason Gunthorpe
2021-04-25 13:03 ` Leon Romanovsky
2021-04-25 13:08 ` Jason Gunthorpe
2021-04-25 13:44 ` Leon Romanovsky
2021-04-25 17:22 ` Jason Gunthorpe
2021-04-25 17:38 ` Leon Romanovsky
2021-04-26 12:03 ` Jason Gunthorpe
2021-04-26 13:08 ` Leon Romanovsky
2021-04-26 13:11 ` Jason Gunthorpe
2021-04-27 4:45 ` Leon Romanovsky
2021-05-02 11:28 ` Leon Romanovsky
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=20210420152541.GC1370958@nvidia.com \
--to=jgg@nvidia.com \
--cc=dledford@redhat.com \
--cc=leon@kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=shayd@nvidia.com \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.