From: Jason Gunthorpe <jgg@nvidia.com>
To: Leon Romanovsky <leon@kernel.org>
Cc: Doug Ledford <dledford@redhat.com>, <linux-rdma@vger.kernel.org>
Subject: Re: [PATCH rdma-next v1 06/10] RDMA/core: Delete function indirection for alloc/free kernel CQ
Date: Thu, 3 Sep 2020 09:24:00 -0300 [thread overview]
Message-ID: <20200903122400.GB1152540@nvidia.com> (raw)
In-Reply-To: <20200903053517.GR59010@unreal>
On Thu, Sep 03, 2020 at 08:35:17AM +0300, Leon Romanovsky wrote:
> On Wed, Sep 02, 2020 at 09:20:48PM -0300, Jason Gunthorpe wrote:
> > On Sun, Aug 30, 2020 at 11:40:06AM +0300, Leon Romanovsky wrote:
> > > /**
> > > - * ib_free_cq_user - free a completion queue
> > > + * ib_free_cq - free a completion queue
> > > * @cq: completion queue to free.
> > > - * @udata: User data or NULL for kernel object
> > > */
> > > -void ib_free_cq_user(struct ib_cq *cq, struct ib_udata *udata)
> > > +void ib_free_cq(struct ib_cq *cq)
> > > {
> > > - if (WARN_ON_ONCE(atomic_read(&cq->usecnt)))
> > > - return;
> > > - if (WARN_ON_ONCE(cq->cqe_used))
> > > - return;
> > > + WARN_ON_ONCE(atomic_read(&cq->usecnt));
> >
> > In this case we expect ops.destroy_cq to fail, so no sense in
> > continuing, leak everything, the ULP is buggy.
>
> I disagree, we should clean as much as possible and leave the HW object.
There is no reason, this just creates possible data corruption.
Once these WARN_ONs are hit, the system is leaking memory. It doesn't
really matter how much is leaked, and there is no reason to try to
minimize the leaking.
Jason
next prev parent reply other threads:[~2020-09-03 12:26 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-30 8:40 [PATCH rdma-next v1 00/10] Restore failure of destroy commands Leon Romanovsky
2020-08-30 8:40 ` [PATCH rdma-next v1 01/10] RDMA: Restore ability to fail on PD deallocate Leon Romanovsky
2020-08-30 8:40 ` [PATCH rdma-next v1 02/10] RDMA: Restore ability to fail on AH destroy Leon Romanovsky
2020-08-30 8:40 ` [PATCH rdma-next v1 03/10] RDMA/mlx5: Issue FW command to destroy SRQ on reentry Leon Romanovsky
2020-09-03 0:31 ` Jason Gunthorpe
2020-09-03 5:08 ` Leon Romanovsky
2020-09-03 11:54 ` Jason Gunthorpe
2020-08-30 8:40 ` [PATCH rdma-next v1 04/10] RDMA/mlx5: Fix potential race between destroy and CQE poll Leon Romanovsky
2020-09-03 13:42 ` Jason Gunthorpe
2020-08-30 8:40 ` [PATCH rdma-next v1 05/10] RDMA: Restore ability to fail on SRQ destroy Leon Romanovsky
2020-09-03 0:08 ` Jason Gunthorpe
2020-09-03 5:11 ` Leon Romanovsky
2020-09-03 11:55 ` Jason Gunthorpe
2020-09-03 0:18 ` Jason Gunthorpe
2020-09-03 5:28 ` Leon Romanovsky
2020-09-03 12:22 ` Jason Gunthorpe
2020-09-03 13:12 ` Jason Gunthorpe
2020-08-30 8:40 ` [PATCH rdma-next v1 06/10] RDMA/core: Delete function indirection for alloc/free kernel CQ Leon Romanovsky
2020-09-03 0:20 ` Jason Gunthorpe
2020-09-03 5:35 ` Leon Romanovsky
2020-09-03 12:24 ` Jason Gunthorpe [this message]
2020-08-30 8:40 ` [PATCH rdma-next v1 07/10] RDMA: Allow fail of destroy CQ Leon Romanovsky
2020-08-30 8:40 ` [PATCH rdma-next v1 08/10] RDMA: Change XRCD destroy return value Leon Romanovsky
2020-08-30 8:40 ` [PATCH rdma-next v1 09/10] RDMA: Restore ability to return error for destroy WQ Leon Romanovsky
2020-08-30 8:40 ` [PATCH rdma-next v1 10/10] RDMA: Make counters destroy symmetrical 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=20200903122400.GB1152540@nvidia.com \
--to=jgg@nvidia.com \
--cc=dledford@redhat.com \
--cc=leon@kernel.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;
as well as URLs for NNTP newsgroup(s).