lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <10056508.nUPlyArG6x@leap>
Date:   Thu, 31 Mar 2022 11:07:45 +0200
From:   "Fabio M. De Francesco" <fmdefrancesco@...il.com>
To:     Dan Carpenter <dan.carpenter@...cle.com>,
        Christoph Hellwig <hch@...radead.org>
Cc:     Wenchao Hao <haowenchao@...wei.com>, axboe@...nel.dk,
        jejb@...ux.ibm.com, linux-block@...r.kernel.org,
        linux-kernel@...r.kernel.org, linux-scsi@...r.kernel.org,
        martin.petersen@...cle.com,
        syzbot+f08c77040fa163a75a46@...kaller.appspotmail.com,
        syzkaller-bugs@...glegroups.com, linfeilong@...wei.com
Subject: Re: [PATCH] scsi: sd: call device_del() if device_add_disk() fails

On giovedì? 31 marzo 2022 07:45:12 CEST Christoph Hellwig wrote:
> > The temptation was to call device_unregister() which is a combined
> > device_del(); device_put(); but when the device_initialize() and
> > device_add() are called separately, then I think it is more readable to
> > call del and put separately as well.
> 
> I think we should also consolidate the initialization side.  Using
> device_register and device_unregister would have prevented this bug
> and I should have switched to that before refactoring the code.
> 
If I don't misunderstand what you wrote, I think you mean something like
the following changes:

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index a390679cf458..7a000a9a9dbe 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -3431,7 +3431,7 @@ static int sd_probe(struct device *dev)
        sdkp->disk_dev.class = &sd_disk_class;
        dev_set_name(&sdkp->disk_dev, "%s", dev_name(dev));
 
-       error = device_add(&sdkp->disk_dev);
+       error = device_register(&sdkp->disk_dev);
        if (error) {
                put_device(&sdkp->disk_dev);
                goto out;
@@ -3474,7 +3474,7 @@ static int sd_probe(struct device *dev)
 
        error = device_add_disk(dev, gd, NULL);
        if (error) {
-               put_device(&sdkp->disk_dev);
+               device_unregister(&sdkp->disk_dev);
                goto out;
        }

@Dan, @Christoph: what do you think of the changes that I've copy-pasted above?

Thanks,

Fabio M. De Francesco



Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ