Message-ID: <3FA8D17D.3060204@gmx.de>
Date: Wed, 05 Nov 2003 11:31:25 +0100
From: "Prakash K. Cheemplavam" <prakashpublic@gmx.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5) Gecko/20031102
MIME-Version: 1.0
To: linux-kernel@vger.kernel.org
Subject: Re: 2.9test9-mm1 and DAO ATAPI cd-burning corrupt
References: <3FA69CDF.5070908@gmx.de> <20031105084007.GZ1477@suse.de> <3FA8C916.3060702@gmx.de> <20031105095457.GG1477@suse.de> <3FA8CA87.2070201@gmx.de> <20031105100120.GH1477@suse.de> <3FA8CCF9.6070700@gmx.de> <20031105101207.GI1477@suse.de> <3FA8CEF1.1050200@gmx.de> <20031105102238.GJ1477@suse.de>
In-Reply-To: <20031105102238.GJ1477@suse.de>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 4229
Lines: 135

Jens Axboe wrote:
> On Wed, Nov 05 2003, Prakash K. Cheemplavam wrote:
> 
>>Jens Axboe wrote:
>>
>>>On Wed, Nov 05 2003, Prakash K. Cheemplavam wrote:
>>>
>>>
>>>>Jens Axboe wrote:
>>>>
>>>>
>>>>>On Wed, Nov 05 2003, Prakash K. Cheemplavam wrote:
>>>>>
>>>>>
>>>>>
>>>>>>Jens Axboe wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>On Wed, Nov 05 2003, Prakash K. Cheemplavam wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>Jens Axboe wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>>On Mon, Nov 03 2003, Prakash K. Cheemplavam wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>>Hi,
>>>>>>>>>>
>>>>>>>>>>well I am using k3b0.10.1 and either choosing cdrdao or cdrecord in 
>>>>>>>>
>>>>>>>>DAO mode to burn the cd ends up in non-bit identical copies, wheres 
>>>>>>>>ion TAO (atleast with my 10x CD-RW I tested) the copy succeded. I 
>>>>>>>>tried several times, and always got this issue.
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>>>bash-2.05b$ md5sum livecd-2.6_10-23-2003.iso
>>>>>>>>>>f73f3a74239dfe94b322b85fd14a306e livecd-2.6_10-23-2003.iso
>>>>>>>>>>
>>>>>>>>>>TAO:
>>>>>>>>>>bash-2.05b$ md5sum /dev/cdroms/cdrom1
>>>>>>>>>>f73f3a74239dfe94b322b85fd14a306e /dev/cdroms/cdrom1
>>>>>>>>>>
>>>>>>>>>>DAO:
>>>>>>>>>>bash-2.05b$ md5sum /dev/cdroms/cdrom1
>>>>>>>>>>09e7e2a51af4c64685831513fbac18c2 /dev/cdroms/cdrom1
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>Could you please try and cmp the two images, finding out how big the
>>>>>>>>>corrupted chunks are and what kind of data they contain?
>>>>>>>>
>>>>>>>>
>>>>>>>>After some further investigation I found out, that the data is NOT 
>>>>>>>>corrupted, but in a way truncated in DAO mode: When I read out the 
>>>>>>>>image 
>>>>>>>
>>>>>>>>from the CD-RW drive, about 5kbyte are missing at the end (and doing 
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>>several burn, it is always the same amount). Strange enough if I read 
>>>>>>>>the DAO burnt disk out by my DVD-ROM, the image can be read out 
>>>>>>>>completely! Can you understand this behaviour? I don't think it is a 
>>>>>>>>problem of my burner, but rather of the atapi driver?
>>>>>>>
>>>>>>>
>>>>>>>Is actual data missing at the end? If yes, I'd say this looks a lot 
>>>>>>>more
>>>>>>>like a cdrecord problem.
>>>>>>
>>>>>>Sorry, I wasn't precise: The data is on the disc, as my DVD-ROM 
>>>>>>restores the full image (md5sum matches), but the CD-RW does not.
>>>>>
>>>>>
>>>>>You need to use the pad option to record.
>>>>
>>>>Uhm, could you be more specific? I just use the k3b frontend to burn, 
>>>>and in the cdrdao manual I couldn't find something useful to it.
>>>
>>>
>>>it's a cdrecord option, I've never used k3b so cannot comment on how to
>>>make it enable that.
>>
>>Hmm, I'll take a look, but I don't really think it is a problem of the 
>>recording programme, otherwise how could my reader read it out completely?
> 
> 
> It isn't a problem of the recorder program. But some drives wont read
> the very end of a disc unless there are some pad blocks at the end.
> Thus, you should always use the cdrecord pad option.

Uhm, ok, I just took a look at the source image and the last (missing) 
4096 bytes are just 00, so nothing critical missing anyway...so your pad 
parameter sounds sensible. :) Should drop a note to the k3b devs then, 
as well...


  > SG_IO or CDROM_SEND_PACKET? The latter should never have been merged in
> cdrecord, and you should never use it. If it is using SG_IO, I'm very
> interested in knowing more. vmstat 1 while doing the burn would be
> interesting, just for 10 seconds where you see the problem.

Uhh, I'll try to find out. I must go now, but I'll report back later.

>>probably is a new issue. Furthermore my hd reading (I once contacted you 
>>because of that) didn't improve with the new kernel, though some 
>>bugfixes in this directions are mentioned by Andrew Morton.
> 
> 
> Don't remember, sorry :)

I probably will make a new topic regarding issues (I think) I found with 
the new mm kernel.

cya,

Prakash

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