Linux SCSI subsystem development
 help / color / mirror / Atom feed
From: Lee Duncan <lduncan@suse.com>
To: Mike Christie <michael.christie@oracle.com>,
	skashyap@marvell.com, cleech@redhat.com, njavali@marvell.com,
	mrangankar@marvell.com, GR-QLogic-Storage-Upstream@marvell.com,
	martin.petersen@oracle.com, linux-scsi@vger.kernel.org,
	jejb@linux.ibm.com
Subject: Re: [PATCH 04/10] scsi: iscsi: Fix endpoint reuse regression
Date: Fri, 8 Apr 2022 10:40:39 -0700	[thread overview]
Message-ID: <f4d0100d-a435-91bf-9f0e-5d3f3c88ec04@suse.com> (raw)
In-Reply-To: <20220408001314.5014-5-michael.christie@oracle.com>

On 4/7/22 17:13, Mike Christie wrote:
> This patch fixes a bug where when using iscsi offload we can free a
> endpoint while userspace still thinks it's active. That then causes the
> endpoint ID to be reused for a new connection's endpoint while userspace
> still thinks the ID is for the original connection. Userspace will then
> end up disconnecting a running connection's endpoint or trying to bind
> to another connection's endpoint.
> 
> This bug is a regression added in:
> 
> Commit 23d6fefbb3f6 ("scsi: iscsi: Fix in-kernel conn failure handling")
> 
> where we added a in kernel ep_disconnect call to fix a bug in:
> 
> Commit 0ab710458da1 ("scsi: iscsi: Perform connection failure entirely in
> kernel space")
> 
> where we would call stop_conn without having done ep_disconnect. This
> early ep_disconnect call will then free the endpoint and it's ID while
> userspace still thinks the ID is valid.
> 
> This patch fixes the early release of the ID by having the in kernel
> recovery code keep a reference to the endpoint until userspace has called
> into the kernel to finish cleaning up the endpoint/connection. It requires
> the previous patch "scsi: iscsi: Release endpoint ID when its freed."
> which moved the freeing of the ID until when the endpoint is released.
> 
> Fixes: 23d6fefbb3f6 ("scsi: iscsi: Fix in-kernel conn failure handling")
> Signed-off-by: Mike Christie <michael.christie@oracle.com>
> ---
>   drivers/scsi/scsi_transport_iscsi.c | 12 +++++++++++-
>   1 file changed, 11 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
> index 1fc7c6bfbd67..f200da049f3b 100644
> --- a/drivers/scsi/scsi_transport_iscsi.c
> +++ b/drivers/scsi/scsi_transport_iscsi.c
> @@ -2247,7 +2247,11 @@ static void iscsi_if_disconnect_bound_ep(struct iscsi_cls_conn *conn,
>   		mutex_unlock(&conn->ep_mutex);
>   
>   		flush_work(&conn->cleanup_work);
> -
> +		/*
> +		 * Userspace is now done with the EP so we can release the ref
> +		 * iscsi_cleanup_conn_work_fn took.
> +		 */
> +		iscsi_put_endpoint(ep);
>   		mutex_lock(&conn->ep_mutex);
>   	}
>   }
> @@ -2322,6 +2326,12 @@ static void iscsi_cleanup_conn_work_fn(struct work_struct *work)
>   		return;
>   	}
>   
> +	/*
> +	 * Get a ref to the ep, so we don't release its ID until after
> +	 * userspace is done referencing it in iscsi_if_disconnect_bound_ep.
> +	 */
> +	if (conn->ep)
> +		get_device(&conn->ep->dev);
>   	iscsi_ep_disconnect(conn, false);
>   
>   	if (system_state != SYSTEM_RUNNING) {

Reviewed-by: Lee Duncan <lduncan@suse.com>


  reply	other threads:[~2022-04-08 17:41 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-08  0:13 [PATCH 00/10] iscsi fixes Mike Christie
2022-04-08  0:13 ` [PATCH 01/10] scsi: iscsi: Move iscsi_ep_disconnect Mike Christie
2022-04-09  1:36   ` Chris Leech
2022-04-08  0:13 ` [PATCH 02/10] scsi: iscsi: Fix offload conn cleanup when iscsid restarts Mike Christie
2022-04-08 17:21   ` Lee Duncan
2022-04-09  1:36   ` Chris Leech
2022-04-08  0:13 ` [PATCH 03/10] scsi: iscsi: Release endpoint ID when its freed Mike Christie
2022-04-08 17:39   ` Lee Duncan
2022-04-09  1:40   ` Chris Leech
2022-04-11  7:22   ` wubo (T)
2022-04-08  0:13 ` [PATCH 04/10] scsi: iscsi: Fix endpoint reuse regression Mike Christie
2022-04-08 17:40   ` Lee Duncan [this message]
2022-04-09  1:41   ` Chris Leech
2022-04-08  0:13 ` [PATCH 05/10] scsi: iscsi: Fix conn cleanup and stop race during iscsid restart Mike Christie
2022-04-08 17:48   ` Lee Duncan
2022-04-09  1:46   ` Chris Leech
2022-04-08  0:13 ` [PATCH 06/10] scsi: iscsi: Fix unbound endpoint error handling Mike Christie
2022-04-08 17:55   ` Lee Duncan
2022-04-09  1:54   ` Chris Leech
2022-04-08  0:13 ` [PATCH 07/10] scsi: iscsi: Merge suspend fields Mike Christie
2022-04-09  1:56   ` Chris Leech
2022-04-08  0:13 ` [PATCH 08/10] scsi: iscsi: Fix nop handling during conn recovery Mike Christie
2022-04-09  1:59   ` Chris Leech
2022-04-08  0:13 ` [PATCH 09/10] scsi: qedi: Fix failed disconnect handling Mike Christie
2022-04-08 16:49   ` [EXT] " Manish Rangankar
2022-04-08 17:58   ` Lee Duncan
2022-04-09  2:00   ` Chris Leech
2022-04-08  0:13 ` [PATCH 10/10] scsi: iscsi: Add Mike Christie as co-maintainer Mike Christie
2022-04-08 17:59   ` Lee Duncan
2022-04-09  1:57   ` Chris Leech
2022-04-08 16:47 ` [EXT] [PATCH 00/10] iscsi fixes Manish Rangankar
2022-04-12  2:36 ` Martin K. Petersen

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=f4d0100d-a435-91bf-9f0e-5d3f3c88ec04@suse.com \
    --to=lduncan@suse.com \
    --cc=GR-QLogic-Storage-Upstream@marvell.com \
    --cc=cleech@redhat.com \
    --cc=jejb@linux.ibm.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=michael.christie@oracle.com \
    --cc=mrangankar@marvell.com \
    --cc=njavali@marvell.com \
    --cc=skashyap@marvell.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox