linux-kernel - Re: [PATCH v2 2/3] usb: gadget: file_storage: Make CD-ROM emulation work with Mac OS-X

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:	Tue, 22 Mar 2011 16:39:12 +0200
From:	Roger Quadros <roger.quadros@...ia.com>
To:	ext Alan Stern <stern@...land.harvard.edu>
CC:	<gregkh@...e.de>, <sshtylyov@...sta.com>,
	<linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2 2/3] usb: gadget: file_storage: Make CD-ROM emulation
 work with Mac OS-X

On 03/22/2011 04:26 PM, ext Alan Stern wrote:
> On Tue, 22 Mar 2011, Roger Quadros wrote:
>
>> I have a question here.
>>
>> The host can request us to send less or more than the actual TOC size, since it
>> has no clue how big it is.
>> e.g. Linux host requests us to send only 12 bytes even though our formatted TOC
>> length is 20. In this case should we return fsg->data_size_from_cmnd instead of
>> actual TOC length?
>
> No.  Always return the actual TOC length.

OK.

>
>> e.g. Mac requests us to send 65534 bytes but our RAW TOC length is 37.
>> The file storage driver seems to be zero padding our data response. So we
>> respond with 65534 bytes, 37 of TOC and remaining zero padded.
>>
>> Can we do something like this to avoid unnecessary zero padded transfers?
>>
>>           ret = fsg_get_toc(curlun, msf, format, buf);
>>           if (ret<  0) {
>>                   curlun->sense_data = SS_INVALID_FIELD_IN_CDB;
>>                   return -EINVAL;
>>           } else if (ret>  fsg->data_size_from_cmnd) {
>>                   ret = fsg->data_size_from_cmnd;
>>           } else {
>>                   fsg->residue = ret;
>>           }
>>           return ret;
>
> Not needed (and not correct).  The code at the end of do_scsi_command()
> already does this:
>
> 		reply = min((u32) reply, fsg->data_size_from_cmnd);
> 		...
> 		fsg->residue -= reply;
>

Oh yes, this explains why to return the actual TOC length in do_read_toc().

However, if host asked for data more than the TOC length then residue will be 
greater than zero and this will result in zero padded transfers.

Not a big issue but should we be fixing it?

one way could be to set fsg->residue to actual TOC length. in do_read_toc().

-- 
regards,
-roger
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/