From: Bart Van Assche <Bart.VanAssche@wdc.com>
To: "jejb@linux.vnet.ibm.com" <jejb@linux.vnet.ibm.com>,
"hare@suse.de" <hare@suse.de>,
"martin.petersen@oracle.com" <martin.petersen@oracle.com>
Cc: "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
"hch@lst.de" <hch@lst.de>,
"jthumshirn@suse.de" <jthumshirn@suse.de>,
"shane.seymour@hpe.com" <shane.seymour@hpe.com>
Subject: Re: [PATCH 07/19] Fix RCU handling of scsi_device.vpd_pg8[03]
Date: Fri, 25 Aug 2017 16:26:28 +0000 [thread overview]
Message-ID: <1503678387.2680.10.camel@wdc.com> (raw)
In-Reply-To: <8abaa903-7150-b7a0-ffc8-2fb4d2695e6c@suse.de>
On Fri, 2017-08-25 at 17:49 +0200, Hannes Reinecke wrote:
> On 08/23/2017 11:39 PM, Bart Van Assche wrote:
> > diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
> > index 3d38c6d463b8..5bb15e698969 100644
> > --- a/drivers/scsi/scsi.c
> > +++ b/drivers/scsi/scsi.c
> > @@ -426,7 +426,7 @@ void scsi_attach_vpd(struct scsi_device *sdev)
> > int vpd_len = SCSI_VPD_PG_LEN;
> > int pg80_supported = 0;
> > int pg83_supported = 0;
> > - unsigned char __rcu *vpd_buf, *orig_vpd_buf = NULL;
> > + unsigned char *vpd_buf, *orig_vpd_buf = NULL;
> >
> > if (!scsi_device_supports_vpd(sdev))
> > return;
>
> Why drop the __rcu annotation here?
> vpd_buf and orig_vpd_buf should always be accessed via rcu pointers.
> Did I misunderstood the meaning of the __rcu annotation?
Hello Hannes,
The __rcu annotation must only be used for pointers that are accessed from
multiple threads. The vpd_buf pointer is a local variable that is only used
from a single context. The family of rcu_dereference() functions expects an
__rcu-annotated pointer as argument and returns a regular pointer. See e.g.
the example in Documentation/RCU/whatisRCU.txt.
> > diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
> > index 5ed473a87589..cf8a2088a9ba 100644
> > --- a/drivers/scsi/scsi_sysfs.c
> > +++ b/drivers/scsi/scsi_sysfs.c
> > @@ -456,8 +456,11 @@ static void scsi_device_dev_release_usercontext(struct work_struct *work)
> > /* NULL queue means the device can't be used */
> > sdev->request_queue = NULL;
> >
> > - kfree(sdev->vpd_pg83);
> > - kfree(sdev->vpd_pg80);
> > + mutex_lock(&sdev->inquiry_mutex);
> > + kfree(rcu_dereference(sdev->vpd_pg83));
> > + kfree(rcu_dereference(sdev->vpd_pg80));
> > + mutex_unlock(&sdev->inquiry_mutex);
> > +
> > kfree(sdev->inquiry);
> > kfree(sdev);
>
> As indicated by Shane, I think using 'kfree_rcu()' here might not be a
> bad idea. You never know ...
I will make that change. Thanks for the reviews!
Bart.
next prev parent reply other threads:[~2017-08-25 16:27 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-23 21:39 [PATCH 00/19] SCSI patches for kernel v4.14 Bart Van Assche
2017-08-23 21:39 ` [PATCH 01/19] Remove an obsolete function declaration Bart Van Assche
2017-08-24 9:02 ` Christoph Hellwig
2017-08-23 21:39 ` [PATCH 02/19] Avoid sign extension of scsi_device.type Bart Van Assche
2017-08-24 6:10 ` Hannes Reinecke
2017-08-24 9:02 ` Christoph Hellwig
2017-08-23 21:39 ` [PATCH 03/19] Suppress gcc 7 fall-through warnings reported with W=1 Bart Van Assche
2017-08-24 9:03 ` Christoph Hellwig
2017-08-23 21:39 ` [PATCH 04/19] Convert a strncmp() call into a strcmp() call Bart Van Assche
2017-08-24 9:03 ` Christoph Hellwig
2017-08-25 15:43 ` Hannes Reinecke
2017-08-23 21:39 ` [PATCH 05/19] scsi_setup_fs_cmnd(): Call scsi_req_init() instead of open-coding it Bart Van Assche
2017-08-24 9:05 ` Christoph Hellwig
2017-08-24 16:17 ` Bart Van Assche
2017-08-23 21:39 ` [PATCH 06/19] Document which queue type a function is intended for Bart Van Assche
2017-08-24 9:05 ` Christoph Hellwig
2017-08-24 16:57 ` Bart Van Assche
2017-08-24 16:58 ` hch
2017-08-24 17:22 ` Bart Van Assche
2017-08-25 15:44 ` Hannes Reinecke
2017-08-23 21:39 ` [PATCH 07/19] Fix RCU handling of scsi_device.vpd_pg8[03] Bart Van Assche
2017-08-24 9:07 ` Christoph Hellwig
2017-08-24 16:54 ` Bart Van Assche
2017-08-25 5:58 ` Seymour, Shane M
2017-08-25 6:59 ` hch
2017-08-25 20:04 ` Bart Van Assche
2017-08-28 2:02 ` Seymour, Shane M
2017-08-25 15:49 ` Hannes Reinecke
2017-08-25 16:26 ` Bart Van Assche [this message]
2017-08-23 21:39 ` [PATCH 08/19] Use blk_mq_rq_to_pdu() to convert a request to a SCSI command pointer Bart Van Assche
2017-08-24 9:07 ` Christoph Hellwig
2017-08-23 21:39 ` [PATCH 09/19] sd, sr: Convert two assignments into warning statements Bart Van Assche
2017-08-24 9:08 ` Christoph Hellwig
2017-08-23 21:40 ` [PATCH 10/19] sd: Fix indentation Bart Van Assche
2017-08-24 9:08 ` Christoph Hellwig
2017-08-25 15:50 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 11/19] sd: Remove a useless comparison Bart Van Assche
2017-08-25 15:50 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 12/19] sg: Fix type of last blk_trace_setup() argument Bart Van Assche
2017-08-24 9:08 ` Christoph Hellwig
2017-08-25 15:51 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 13/19] libiscsi: Fix indentation Bart Van Assche
2017-08-24 9:08 ` Christoph Hellwig
2017-08-25 15:51 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 14/19] libsas: Remove a set-but-not-used variable Bart Van Assche
2017-08-24 9:09 ` Christoph Hellwig
2017-08-25 15:51 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 15/19] libsas: Annotate fall-through in a switch statement Bart Van Assche
2017-08-24 9:09 ` Christoph Hellwig
2017-08-25 15:52 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 16/19] scsi_transport_sas, sas_tlr_supported(): Check kzalloc() return value Bart Van Assche
2017-08-24 9:09 ` Christoph Hellwig
2017-08-25 15:52 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 17/19] scsi_transport_srp: Suppress a W=1 compiler warning Bart Van Assche
2017-08-24 9:11 ` Christoph Hellwig
2017-08-24 16:27 ` Bart Van Assche
2017-08-25 15:29 ` hch
2017-08-25 15:40 ` Bart Van Assche
2017-08-25 15:56 ` hch
2017-08-23 21:40 ` [PATCH 18/19] scsi_debug: Remove a set-but-not-used variable Bart Van Assche
2017-08-24 9:12 ` Christoph Hellwig
2017-08-25 15:53 ` Hannes Reinecke
2017-08-23 21:40 ` [PATCH 19/19] iscsi_tcp: " Bart Van Assche
2017-08-24 9:12 ` Christoph Hellwig
2017-08-25 15:53 ` Hannes Reinecke
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=1503678387.2680.10.camel@wdc.com \
--to=bart.vanassche@wdc.com \
--cc=hare@suse.de \
--cc=hch@lst.de \
--cc=jejb@linux.vnet.ibm.com \
--cc=jthumshirn@suse.de \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=shane.seymour@hpe.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