linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Christian Löhle" <CLoehle@hyperstone.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
	"axboe@kernel.dk" <axboe@kernel.dk>
Subject: Re: [PATCH] block: prevent sending events from removed device
Date: Thu, 1 Jul 2021 08:23:39 +0000	[thread overview]
Message-ID: <CWXP265MB2680A529C732A3137C15991DC4009@CWXP265MB2680.GBRP265.PROD.OUTLOOK.COM> (raw)
In-Reply-To: <YN1xzGPbx8ac8r3j@infradead.org>

Alright Thanks, I will try out your suggestion and report back.
Regards,
Christian

>I don't see how this is going to work.  If the device is being
>deleted the reference count will be zero and refcount_inc as
>called from kobject_get will WARN.  We'll need to check the
>disk is marked up, and we need to do that under a lock.
>
>Something like this completely untested patch:

diff --git a/block/disk-events.c b/block/disk-events.c
index a75931ff5da4..27b845c51f2a 100644
--- a/block/disk-events.c
+++ b/block/disk-events.c
@@ -190,6 +190,9 @@ static void disk_check_events(struct disk_events *ev,
 
         spin_unlock_irq(&ev->lock);
 
+       if (!(disk->flags & GENHD_FL_UP))
+               return;
+
         /*
          * Tell userland about new events.  Only the events listed in
          * @disk->events are reported, and only if DISK_EVENT_FLAG_UEVENT
@@ -268,6 +271,8 @@ bool bdev_check_media_change(struct block_device *bdev)
 {
         unsigned int events;
 
+       lockdep_assert_held(&bdev->bd_disk->open_mutex);
+
         events = disk_clear_events(bdev->bd_disk, DISK_EVENT_MEDIA_CHANGE |
                                    DISK_EVENT_EJECT_REQUEST);
         if (!(events & DISK_EVENT_MEDIA_CHANGE))
@@ -290,7 +295,10 @@ static void disk_events_workfn(struct work_struct *work)
         struct delayed_work *dwork = to_delayed_work(work);
         struct disk_events *ev = container_of(dwork, struct disk_events, dwork);
 
-       disk_check_events(ev, &ev->clearing);
+       mutex_lock(&ev->disk->open_mutex);
+       if (ev->disk->flags & GENHD_FL_UP)
+               disk_check_events(ev, &ev->clearing);
+       mutex_unlock(&ev->disk->open_mutex);
 }
 
 /*
diff --git a/drivers/block/ataflop.c b/drivers/block/ataflop.c
index a093644ac39f..b8e77da44235 100644
--- a/drivers/block/ataflop.c
+++ b/drivers/block/ataflop.c
@@ -1735,8 +1735,10 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode,
                 /* invalidate the buffer track to force a reread */
                 BufferDrive = -1;
                 set_bit(drive, &fake_change);
+               mutex_lock(&bdev->bd_disk->open_mutex);
                 if (bdev_check_media_change(bdev))
                         floppy_revalidate(bdev->bd_disk);
+               mutex_unlock(&bdev->bd_disk->open_mutex);
                 return 0;
         default:
                 return -EINVAL;
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 87460e0e5c72..2a97f22cfa0b 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -3185,8 +3185,10 @@ static int invalidate_drive(struct block_device *bdev)
         /* invalidate the buffer track to force a reread */
         set_bit((long)bdev->bd_disk->private_data, &fake_change);
         process_fd_request();
+       mutex_lock(&bdev->bd_disk->open_mutex);
         if (bdev_check_media_change(bdev))
                 floppy_revalidate(bdev->bd_disk);
+       mutex_unlock(&bdev->bd_disk->open_mutex);
         return 0;
 }
 

Hyperstone GmbH | Line-Eid-Strasse 3 | 78467 Konstanz
Managing Directors: Dr. Jan Peter Berns.
Commercial register of local courts: Freiburg HRB381782


      reply	other threads:[~2021-07-01  8:23 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-30  8:09 [PATCH] block: prevent sending events from removed device Christian Löhle
2021-06-30 11:46 ` [PATCHv2] " Christian Löhle
2021-07-01  7:42 ` [PATCH] " Christoph Hellwig
2021-07-01  8:23   ` Christian Löhle [this message]

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=CWXP265MB2680A529C732A3137C15991DC4009@CWXP265MB2680.GBRP265.PROD.OUTLOOK.COM \
    --to=cloehle@hyperstone.com \
    --cc=axboe@kernel.dk \
    --cc=hch@infradead.org \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@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).