Re: Ext. HDDs not working under Linux via USB 3.0 but under Windows

From: "Till Dörges" <doerges@pre-sense.de>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: linux-usb@vger.kernel.org
Subject: Re: Ext. HDDs not working under Linux via USB 3.0 but under Windows
Date: Thu, 18 Jun 2020 08:07:55 +0200	[thread overview]
Message-ID: <763037ac-d8fa-8697-7f75-c0bf958b4308@pre-sense.de> (raw)
In-Reply-To: <20200617163734.GC11314@rowland.harvard.edu>

Am 17.06.20 um 18:37 schrieb Alan Stern:
> On Wed, Jun 17, 2020 at 06:02:32PM +0200, Till Dörges wrote:
>> Hi Alan,
>>
>> sorry for the delay. It took me a while to get back to this.
>>
>> Am 23.05.20 um 22:33 schrieb Alan Stern:
>>
>>>>>> I'm trying to get external HDDs from Sony to work under Linux with USB 3.0.
>>>> [...]
>>>>>>  - PSZ-HA2T: idVendor=054c, idProduct=087d
>>>>>>  - PSZ-HA1T: idVendor=054c, idProduct=087d
>>>>>>  - PSZ-HA50: idVendor=054c, idProduct=087d
>>>>>>  - PSZ-HC1T: idVendor=054c, idProduct=0c4f
>>>> [...]
>>>>>> I've also tried several of the quirk settings
>>>>>> (https://github.com/torvalds/linux/blob/ead751507de86d90fa250431e9990a8b881f713c/drivers/usb/storage/usb.c#L527).
>>
>> [...]
>>>>> Try adding an entry to the unusual_uas.h file for the device, specifying 
>>>>> the USB_FL_NO_REPORT_OPCODES flag.
>>>>
>>>> I'm not sure I understand correctly. Do I have to compile the uas Kernel module
>>>> myself if I want to test a flag?
>>>>
>>>> Because with setting the f flag (via options usb-storage quirks=054c:087d:f) the
>>>> PSZ-HA*1*T starts behaving nicely.
>>>>
>>>> --- snip ---
>>>> $ cat /sys/module/usb_storage/parameters/quirks
>>>> 054c:087d:f,054c:0c4f:f
>>>> --- snip ---
>>>>
>>>>
>>>> The PSZ-HA*2*T, however, still doesn't work with that.
>> Since the device stopped working under windows, I'm guessing that it's simply broken.
>>
>> I'll try to get my hands on a new sample.
>>
>>
>> The PSZ-HA50 and the PSZ-HA1T work nicely when attached to USB 3.0 with this setting:
>>
>> --- snip ---
>> $ cat /sys/module/usb_storage/parameters/quirks
>> 054c:087d:u
>> --- snip ---
>>
>> W/o this quirk they do not work.
>>
>>
>> I'd be surprised if a non-broken PSZ-HA2T behaved any differently.
>>
>> So perhaps you might want to consider adding that quirk directly into the Kernel?
>>
>>
>> FTR, the PSZ-HC1T just worked.
> 
> Okay.  If you would like to write a patch for the quirk and submit it, you 
> can.  Otherwise I'll write one for you.

I'd gladly take your offer of writing one for me. :-)

Thanks -- Till
-- 
Dipl.-Inform. Till Dörges                  doerges@pre-sense.de
PRESENSE Technologies GmbH             Nagelsweg 41, D-20097 HH
Geschäftsführer/Managing Directors       AG Hamburg, HRB 107844
Till Dörges, Jürgen Sander               USt-IdNr.: DE263765024