From: Leon Romanovsky <leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
To: Jason Gunthorpe <jgg-uk2M96/98Pc@public.gmane.org>
Cc: Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Mark Bloch <markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
Subject: Re: [PATCH rdma-next 6/8] RDMA/netlink: Protect device query from device removal
Date: Thu, 28 Dec 2017 07:04:54 +0200 [thread overview]
Message-ID: <20171228050454.GK3494@mtr-leonro.local> (raw)
In-Reply-To: <20171227230552.GL25436-uk2M96/98Pc@public.gmane.org>
[-- Attachment #1: Type: text/plain, Size: 1887 bytes --]
On Wed, Dec 27, 2017 at 04:05:52PM -0700, Jason Gunthorpe wrote:
> On Sun, Dec 24, 2017 at 03:43:26PM +0200, Leon Romanovsky wrote:
> > From: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> >
> > There is a chance that device will be removed during device query
> > operations and it will cause to kernel panic in the flows which
> > doesn't hold lists_rwsem semaphore.
> >
> > Fixes: e5c9469efcb1 ("RDMA/netlink: Add nldev device doit implementation")
> > Reviewed-by: Mark Bloch <markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > Signed-off-by: Leon Romanovsky <leonro-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
> > drivers/infiniband/core/nldev.c | 25 +++++++++++++++----------
> > 1 file changed, 15 insertions(+), 10 deletions(-)
> >
> > diff --git a/drivers/infiniband/core/nldev.c b/drivers/infiniband/core/nldev.c
> > index 2b631307349d..e3033d7a4029 100644
> > +++ b/drivers/infiniband/core/nldev.c
> > @@ -141,36 +141,41 @@ static int nldev_get_doit(struct sk_buff *skb, struct nlmsghdr *nlh,
> > struct ib_device *device;
> > struct sk_buff *msg;
> > u32 index;
> > - int err;
> > + int ret = -ENOMEM;
> >
> > - err = nlmsg_parse(nlh, 0, tb, RDMA_NLDEV_ATTR_MAX - 1,
> > + ret = nlmsg_parse(nlh, 0, tb, RDMA_NLDEV_ATTR_MAX - 1,
> > nldev_policy, extack);
> > - if (err || !tb[RDMA_NLDEV_ATTR_DEV_INDEX])
> > + if (ret || !tb[RDMA_NLDEV_ATTR_DEV_INDEX])
> > return -EINVAL;
> >
> > index = nla_get_u32(tb[RDMA_NLDEV_ATTR_DEV_INDEX]);
> >
> > - device = __ib_device_get_by_index(index);
>
> It is not possible to correctly use __ib_device_get_by_index without
> grabbing one of the locks which are static to device.c
>
> Thus it should not be in core_priv.h and every single user must be
> wrong, please fix them all here and make it static again.
Ok, I'll squash patch 6 & 7 and remove the declaration from core_priv.h
Thanks
>
> Jason
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2017-12-28 5:04 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-12-24 13:43 [PATCH rdma-next 0/8] RDMA fixes and refactoring rdma-next-2017-12-24 Leon Romanovsky
[not found] ` <20171224134328.17398-1-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2017-12-24 13:43 ` [PATCH rdma-next 1/8] RDMA/rxe: Remove useless export symbol Leon Romanovsky
2017-12-24 13:43 ` [PATCH rdma-next 2/8] RDMA/netlink: Simplify code of autoload modules Leon Romanovsky
2017-12-24 13:43 ` [PATCH rdma-next 3/8] RDMA/core: Replace open-coded variant of put_device Leon Romanovsky
2017-12-24 13:43 ` [PATCH rdma-next 4/8] RDMA/nldev: Refactor nldev handle to be common function Leon Romanovsky
2017-12-24 13:43 ` [PATCH rdma-next 5/8] RDMA/core: Provide locked variant of device name to index function Leon Romanovsky
[not found] ` <20171224134328.17398-6-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2017-12-27 23:02 ` Jason Gunthorpe
[not found] ` <20171227230227.GK25436-uk2M96/98Pc@public.gmane.org>
2017-12-28 5:01 ` Leon Romanovsky
2017-12-24 13:43 ` [PATCH rdma-next 6/8] RDMA/netlink: Protect device query from device removal Leon Romanovsky
[not found] ` <20171224134328.17398-7-leon-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2017-12-27 23:05 ` Jason Gunthorpe
[not found] ` <20171227230552.GL25436-uk2M96/98Pc@public.gmane.org>
2017-12-28 5:04 ` Leon Romanovsky [this message]
2017-12-24 13:43 ` [PATCH rdma-next 7/8] RDMA/nldev: Protect port query from accidental " Leon Romanovsky
2017-12-24 13:43 ` [PATCH rdma-next 8/8] RDMA/cma: Mark end of CMA ID messages 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=20171228050454.GK3494@mtr-leonro.local \
--to=leon-dgejt+ai2ygdnm+yrofe0a@public.gmane.org \
--cc=dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=jgg-uk2M96/98Pc@public.gmane.org \
--cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=markb-VPRAkNaXOzVWk0Htik3J/w@public.gmane.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