All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgg@nvidia.com>
To: Leon Romanovsky <leon@kernel.org>
Cc: Doug Ledford <dledford@redhat.com>,
	Mark Zhang <markzhang@nvidia.com>,
	linux-rdma@vger.kernel.org
Subject: Re: [PATCH rdma-next v2 8/9] IB/cm: Add lock protection when access av/alt_av's port of a cm_id
Date: Thu, 22 Apr 2021 16:08:14 -0300	[thread overview]
Message-ID: <20210422190814.GA2451164@nvidia.com> (raw)
In-Reply-To: <a50fca26e37799491778e5efbf6b6ef21f1c3fbe.1619004798.git.leonro@nvidia.com>

On Wed, Apr 21, 2021 at 02:40:38PM +0300, Leon Romanovsky wrote:
> @@ -303,20 +304,37 @@ static struct ib_mad_send_buf *cm_alloc_msg(struct cm_id_private *cm_id_priv)
>  	struct ib_mad_agent *mad_agent;
>  	struct ib_mad_send_buf *m;
>  	struct ib_ah *ah;
> +	int ret;
> +
> +	read_lock(&cm_id_priv->av_rwlock);
> +	if (!cm_id_priv->av.port) {
> +		ret = -EINVAL;
> +		goto out;
> +	}
>  
>  	mad_agent = cm_id_priv->av.port->mad_agent;
> +	if (!mad_agent) {
> +		ret = -EINVAL;
> +		goto out;
> +	}
> +
>  	ah = rdma_create_ah(mad_agent->qp->pd, &cm_id_priv->av.ah_attr, 0);
> -	if (IS_ERR(ah))
> -		return (void *)ah;
> +	if (IS_ERR(ah)) {
> +		ret = PTR_ERR(ah);
> +		goto out;
> +	}
>  
>  	m = ib_create_send_mad(mad_agent, cm_id_priv->id.remote_cm_qpn,
>  			       cm_id_priv->av.pkey_index,
>  			       0, IB_MGMT_MAD_HDR, IB_MGMT_MAD_DATA,
>  			       GFP_ATOMIC,
>  			       IB_MGMT_BASE_VERSION);
> +
> +	read_unlock(&cm_id_priv->av_rwlock);
>  	if (IS_ERR(m)) {
>  		rdma_destroy_ah(ah, 0);
> -		return m;
> +		ret = PTR_ERR(m);
> +		goto out;
>  	}
>  
>  	/* Timeout set by caller if response is expected. */
> @@ -326,6 +344,10 @@ static struct ib_mad_send_buf *cm_alloc_msg(struct cm_id_private *cm_id_priv)
>  	refcount_inc(&cm_id_priv->refcount);
>  	m->context[0] = cm_id_priv;
>  	return m;
> +
> +out:
> +	read_unlock(&cm_id_priv->av_rwlock);

This flow has read_unlock happening twice on error

Jason

  reply	other threads:[~2021-04-22 19:08 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-21 11:40 [PATCH rdma-next v2 0/9] Fix memory corruption in CM Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 1/9] IB/cm: Pair cm_alloc_response_msg() with a cm_free_response_msg() Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 2/9] IB/cm: Split cm_alloc_msg() Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 3/9] IB/cm: Call the correct message free functions in cm_send_handler() Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 4/9] IB/cm: Tidy remaining cm_msg free paths Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 5/9] Revert "IB/cm: Mark stale CM id's whenever the mad agent was unregistered" Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 6/9] IB/cm: Simplify ib_cancel_mad() and ib_modify_mad() calls Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 7/9] IB/cm: Clear all associated AV's ports when remove a cm device Leon Romanovsky
2021-04-22 19:34   ` Jason Gunthorpe
2021-04-23 13:14     ` Mark Zhang
2021-04-23 14:24       ` Jason Gunthorpe
2021-04-24  2:33         ` Mark Zhang
2021-04-26 13:56           ` Jason Gunthorpe
2021-04-27  1:59             ` Mark Zhang
2021-04-21 11:40 ` [PATCH rdma-next v2 8/9] IB/cm: Add lock protection when access av/alt_av's port of a cm_id Leon Romanovsky
2021-04-22 19:08   ` Jason Gunthorpe [this message]
2021-04-25 13:21     ` Leon Romanovsky
2021-04-21 11:40 ` [PATCH rdma-next v2 9/9] IB/cm: Initialize av before aquire the spin lock in cm_lap_handler 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=20210422190814.GA2451164@nvidia.com \
    --to=jgg@nvidia.com \
    --cc=dledford@redhat.com \
    --cc=leon@kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=markzhang@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.