linux-kernel - Re: [PATCH v2] USB:UAS:return ENODEV when submit urbs fail with device not attached.

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]

Date: Wed, 28 Feb 2024 15:47:51 +0100
From: Oliver Neukum <oneukum@...e.com>
To: "WeitaoWang-oc@...oxin.com" <WeitaoWang-oc@...oxin.com>,
 Oliver Neukum <oneukum@...e.com>, stern@...land.harvard.edu,
 gregkh@...uxfoundation.org, linux-usb@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-scsi@...r.kernel.org,
 usb-storage@...ts.one-eyed-alien.net
Cc: WeitaoWang@...oxin.com, stable@...r.kernel.org
Subject: Re: [PATCH v2] USB:UAS:return ENODEV when submit urbs fail with
 device not attached.

On 28.02.24 23:32, WeitaoWang-oc@...oxin.com wrote:

> @@ -602,6 +606,8 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
>           if (err) {
>               usb_unanchor_urb(cmdinfo->data_out_urb);
>               uas_log_cmd_state(cmnd, "data out submit err", err);
> +            if (err == -ENODEV)
> +                return -ENODEV;

This is a generic error code from errno.h

>               return SCSI_MLQUEUE_DEVICE_BUSY;

This is not.

>           }
>           cmdinfo->state &= ~SUBMIT_DATA_OUT_URB;
> @@ -621,6 +627,8 @@ static int uas_submit_urbs(struct scsi_cmnd *cmnd,
>           if (err) {
>               usb_unanchor_urb(cmdinfo->cmd_urb);
>               uas_log_cmd_state(cmnd, "cmd submit err", err);
> +            if (err == -ENODEV)
> +                return -ENODEV;
>               return SCSI_MLQUEUE_DEVICE_BUSY;
>           }
> 
> I'm not sure I fully understand what your mean.
> Whether the above code is more reasonable? If not,could you give me some
> suggestion? Thanks for your help!

You want to change uas_submit_urbs() to return the reason for
errors, because -ENODEV needs to be handled differently. That
is good.
But why don't you just do

return err;

unconditionally? There is no point in using SCSI_MLQUEUE_DEVICE_BUSY

	Regards
		Oliver