[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <8de18ee2-ccdd-4cdd-ae49-48600ad30ed4@suse.com>
Date: Mon, 27 Oct 2025 14:35:37 +0100
From: Oliver Neukum <oneukum@...e.com>
To: Owen Gu <guhuinan@...omi.com>, Oliver Neukum <oneukum@...e.com>,
Alan Stern <stern@...land.harvard.edu>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc: linux-usb@...r.kernel.org, linux-scsi@...r.kernel.org,
usb-storage@...ts.one-eyed-alien.net, linux-kernel@...r.kernel.org,
Yu Chen <chenyu45@...omi.com>, Michal Pecio <michal.pecio@...il.com>
Subject: Re: [PATCH v2] usb: uas: fix urb unmapping issue when the uas device
is remove during ongoing data transfer
Hi,
I think I was unclear the first time.
Sorry for that.
On 27.10.25 07:05, Owen Gu wrote:
> Hi Oliver,
>
>> This patch modifies the error condition check in the uas_submit_urbs()
>> function. When a UAS device is removed but one or more URBs have already
>> been successfully submitted to USB, it avoids immediately invoking
>> scsi_done() and save the cmnd to devinfo->cmnd array. If the successfully
>> submitted URBs is completed before devinfo->resetting being set, then
>> the scsi_done() function will be called within uas_try_complete() after
This requires that uas_try_complete() is called.
And I am afraid uas_stat_cmplt() cannot guarantee that in the error case.
I think the following sequence of events is possible:
CPU A CPU B
uas_queuecommand_lck() calls uas_submit_urbs()
COMMAND_INFLIGHT is set and URB submitted
URB gets an error
status = -EBABBLE (just an example)
uas_stat_cmplt is called
resetting is not set
if (status)
goto out;
uas_try_complete _not_ called
The scsi command runs for indeterminate amount of time.
It seems to me that if you want to use your approach you also
need to change error handling in uas_stat_cmplt()
Regards
Oliver
Powered by blists - more mailing lists