[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d656179616d61623e1c2b45c1df219c0693749a1.camel@mediatek.com>
Date: Fri, 15 Oct 2021 21:33:08 +0800
From: Stanley Chu <stanley.chu@...iatek.com>
To: Miles Chen <miles.chen@...iatek.com>,
"James E . J . Bottomley" <jejb@...ux.ibm.com>,
"Martin K . Petersen" <martin.petersen@...cle.com>,
Matthias Brugger <matthias.bgg@...il.com>,
Bart Van Assche <bvanassche@....org>,
Martin Kepplinger <martink@...teo.de>
CC: <linux-scsi@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>,
<linux-mediatek@...ts.infradead.org>, <wsd_upstream@...iatek.com>
Subject: Re: [PATCH] scsi: sd: fix crashes in sd_resume_runtime
Hi Miles,
On Fri, 2021-10-15 at 15:46 +0800, Miles Chen wrote:
> After merging commit ed4246d37f3b ("scsi: sd: REQUEST SENSE for
> BLIST_IGN_MEDIA_CHANGE devices in runtime_resume()"), I hit the
> following crash on my device.
>
> static int sd_resume_runtime(struct device *dev)
> {
> struct scsi_disk *sdkp = dev_get_drvdata(dev);
> struct scsi_device *sdp = sdkp->device; // sdkp == NULL and
> crash
>
> if (sdp->ignore_media_change) {
> ...
> }
>
> I checked sd_resume() and found that sdkp is possbile to be NULL, and
> there is a null pointer test in sd_resume() for this case.
> To fix this crash, follow sd_resume() to test if sdkp is NULL
> before dereferencing it.
LGTM.
Reviewed-by: Stanley Chu <stanley.chu@...iatek.com>
Powered by blists - more mailing lists