no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Thomas Voegtle]

Hello,
  
cdrecord -dev=ATAPI -scanbus  with 2.6.0-test11-bk10 and bk13 shows this:
  
scsibus0:
        0,0,0     0) '' '' '' NON CCS Disk
        0,1,0     1) '' '' '' NON CCS Disk

  
but this works well with 2.6.0-test11.
=>
  
        0,0,0     0) 'CREATIVE' ' CD5233E        ' '2.05' Removable CD-ROM
        0,1,0     1) 'PLEXTOR ' 'CD-R   PX-W1610A' '1.04' Removable CD-ROM
  
SuSE 9.0
  
please cc me, I'm not subscribed
  
Greetings
Thomas



-- 
 Thomas Vögtle    email: thomas@voegtle-clan.de
 ----- http://www.voegtle-clan.de/thomas ------

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Måns Rullgård]

Thomas Voegtle <thomas@voegtle-clan.de> writes:

> cdrecord -dev=ATAPI -scanbus  with 2.6.0-test11-bk10 and bk13 shows this:

cdrecord dev=/dev/cdrom -scanbus  is the recommended way, whatever
cdrecord tries to make you believe.

-- 
Måns Rullgård
mru@kth.se

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Thomas Voegtle]

On Wed, 17 Dec 2003, Måns Rullgård wrote:

> The following message is a courtesy copy of an article
> that has been posted to gmane.linux.kernel as well.
> 
> Thomas Voegtle <thomas@voegtle-clan.de> writes:
> 
> > cdrecord -dev=ATAPI -scanbus  with 2.6.0-test11-bk10 and bk13 shows this:
> 
> cdrecord dev=/dev/cdrom -scanbus  is the recommended way, whatever
> cdrecord tries to make you believe.

Well, this doesn't change the fact that I cannot burn cdroms.
But I can burn with vanilla 2.6.0-test11

Greetings


-- 
 Thomas Vögtle    email: thomas@voegtle-clan.de
 ----- http://www.voegtle-clan.de/thomas ------

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Martin Schlemmer]

[-- Attachment #1: Type: text/plain, Size: 433 bytes --]

On Wed, 2003-12-17 at 17:20, Måns Rullgård wrote:
> Thomas Voegtle <thomas@voegtle-clan.de> writes:
> 
> > cdrecord -dev=ATAPI -scanbus  with 2.6.0-test11-bk10 and bk13 shows this:
> 
> cdrecord dev=/dev/cdrom -scanbus  is the recommended way, whatever
> cdrecord tries to make you believe.

Well, it might be related to my problem of k3b not being able
to detect any devices since bk9 at least.


-- 
Martin Schlemmer

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Gene Heskett]

On Wednesday 17 December 2003 10:09, Thomas Voegtle wrote:
>Hello,
>
>cdrecord -dev=ATAPI -scanbus  with 2.6.0-test11-bk10 and bk13 shows
> this:
>
>scsibus0:
>        0,0,0     0) '' '' '' NON CCS Disk
>        0,1,0     1) '' '' '' NON CCS Disk
>
>
>but this works well with 2.6.0-test11.
>=>
>
>        0,0,0     0) 'CREATIVE' ' CD5233E        ' '2.05' Removable
> CD-ROM 0,1,0     1) 'PLEXTOR ' 'CD-R   PX-W1610A' '1.04' Removable
> CD-ROM
>
>SuSE 9.0

The above looks ok, but even though I'm working with 2.6.0-test11, 
that command gets this:

[root@coyote amanda-dbg]# cdrecord -dev=ATAPI -scanbus
Cdrecord 2.00.3 (i686-pc-linux-gnu) Copyright (C) 1995-2002 Jörg 
Schilling
scsidev: 'ATAPI'
devname: 'ATAPI'
scsibus: -2 target: -2 lun: -2
Warning: Using ATA Packet interface.
Warning: The related libscg interface code is in pre alpha.
Warning: There may be fatal problems.
Using libscg version 'schily-0.7'
scsibus0:
        0,0,0     0) 'CREATIVE' 'CD-RW RW1210E   ' 'LCS6' Removable 
CD-ROM
        0,1,0     1) *
        0,2,0     2) *
        0,3,0     3) *
        0,4,0     4) *
        0,5,0     5) *
        0,6,0     6) *
        0,7,0     7) *

I take that it is attempting to scan all 8 addresses of the scsi bus 
even though its actually hitting the atapi stuff?  Or do I need an 
even fresher version of cdrecord? or libscg?

>please cc me, I'm not subscribed
>
>Greetings
>Thomas

-- 
Cheers, Gene
AMD K6-III@500mhz 320M
Athlon1600XP@1400mhz  512M
99.22% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attornies please note, additions to this message
by Gene Heskett are:
Copyright 2003 by Maurice Eugene Heskett, all rights reserved.

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Thomas Voegtle]

On Wed, 17 Dec 2003, Gene Heskett wrote:

> I take that it is attempting to scan all 8 addresses of the scsi bus 
> even though its actually hitting the atapi stuff?  Or do I need an 
> even fresher version of cdrecord? or libscg?

Sorry, I shortend my output of cdrecord. With 2.6.0-test11 it looks like
this:

Using libscg version 'schily-0.7'
scsibus0:
cdrecord: Warning: controller returns wrong size for CD capabilities page.
        0,0,0     0) 'CREATIVE' ' CD5233E        ' '2.05' Removable CD-ROM
        0,1,0     1) 'PLEXTOR ' 'CD-R   PX-W1610A' '1.04' Removable CD-ROM
        0,2,0     2) *
        0,3,0     3) *
        0,4,0     4) *
        0,5,0     5) *
        0,6,0     6) *
        0,7,0     7) *



-- 
 Thomas Vögtle    email: thomas@voegtle-clan.de
 ----- http://www.voegtle-clan.de/thomas ------

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Gene Heskett]

On Wednesday 17 December 2003 11:25, Thomas Voegtle wrote:
>On Wed, 17 Dec 2003, Gene Heskett wrote:
>> I take that it is attempting to scan all 8 addresses of the scsi
>> bus even though its actually hitting the atapi stuff?  Or do I
>> need an even fresher version of cdrecord? or libscg?
>
>Sorry, I shortend my output of cdrecord. With 2.6.0-test11 it looks
> like this:
>
>Using libscg version 'schily-0.7'
>scsibus0:
>cdrecord: Warning: controller returns wrong size for CD capabilities
> page. 0,0,0     0) 'CREATIVE' ' CD5233E        ' '2.05' Removable
> CD-ROM 0,1,0     1) 'PLEXTOR ' 'CD-R   PX-W1610A' '1.04' Removable
> CD-ROM 0,2,0     2) *
>        0,3,0     3) *
>        0,4,0     4) *
>        0,5,0     5) *
>        0,6,0     6) *
>        0,7,0     7) *

I see.  I also don't see the warning you are getting, and this may be 
the reason you can't burn.  I also do not have the normal cdrom as 
device 0.  What happens if you swap the master/slave jumpers and put 
the recorder first?  It might be worth a try, and any changes in how 
it works would be a clue as to where the real stoppage is.

-- 
Cheers, Gene
AMD K6-III@500mhz 320M
Athlon1600XP@1400mhz  512M
99.22% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attornies please note, additions to this message
by Gene Heskett are:
Copyright 2003 by Maurice Eugene Heskett, all rights reserved.

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Jens Axboe]

On Wed, Dec 17 2003, Gene Heskett wrote:
> On Wednesday 17 December 2003 11:25, Thomas Voegtle wrote:
> >On Wed, 17 Dec 2003, Gene Heskett wrote:
> >> I take that it is attempting to scan all 8 addresses of the scsi
> >> bus even though its actually hitting the atapi stuff?  Or do I
> >> need an even fresher version of cdrecord? or libscg?
> >
> >Sorry, I shortend my output of cdrecord. With 2.6.0-test11 it looks
> > like this:
> >
> >Using libscg version 'schily-0.7'
> >scsibus0:
> >cdrecord: Warning: controller returns wrong size for CD capabilities
> > page. 0,0,0     0) 'CREATIVE' ' CD5233E        ' '2.05' Removable
> > CD-ROM 0,1,0     1) 'PLEXTOR ' 'CD-R   PX-W1610A' '1.04' Removable
> > CD-ROM 0,2,0     2) *
> >        0,3,0     3) *
> >        0,4,0     4) *
> >        0,5,0     5) *
> >        0,6,0     6) *
> >        0,7,0     7) *
> 
> I see.  I also don't see the warning you are getting, and this may be 
> the reason you can't burn.  I also do not have the normal cdrom as 
> device 0.  What happens if you swap the master/slave jumpers and put 
> the recorder first?  It might be worth a try, and any changes in how 
> it works would be a clue as to where the real stoppage is.

The reason is that Thomas is using ATAPI which will go through
CDROM_SEND_PACKET (which is broken in -bk), while you are probably not
using that transport (and thus going to SG_IO directly, which works).

-- 
Jens Axboe

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Gene Heskett]

On Wednesday 17 December 2003 11:49, Jens Axboe wrote:
>On Wed, Dec 17 2003, Gene Heskett wrote:
>> On Wednesday 17 December 2003 11:25, Thomas Voegtle wrote:
>> >On Wed, 17 Dec 2003, Gene Heskett wrote:
>> >> I take that it is attempting to scan all 8 addresses of the
>> >> scsi bus even though its actually hitting the atapi stuff?  Or
>> >> do I need an even fresher version of cdrecord? or libscg?
>> >
>> >Sorry, I shortend my output of cdrecord. With 2.6.0-test11 it
>> > looks like this:
>> >
>> >Using libscg version 'schily-0.7'
>> >scsibus0:
>> >cdrecord: Warning: controller returns wrong size for CD
>> > capabilities page. 0,0,0     0) 'CREATIVE' ' CD5233E        '
>> > '2.05' Removable CD-ROM 0,1,0     1) 'PLEXTOR ' 'CD-R  
>> > PX-W1610A' '1.04' Removable CD-ROM 0,2,0     2) *
>> >        0,3,0     3) *
>> >        0,4,0     4) *
>> >        0,5,0     5) *
>> >        0,6,0     6) *
>> >        0,7,0     7) *
>>
>> I see.  I also don't see the warning you are getting, and this may
>> be the reason you can't burn.  I also do not have the normal cdrom
>> as device 0.  What happens if you swap the master/slave jumpers
>> and put the recorder first?  It might be worth a try, and any
>> changes in how it works would be a clue as to where the real
>> stoppage is.
>
>The reason is that Thomas is using ATAPI which will go through
>CDROM_SEND_PACKET (which is broken in -bk), while you are probably
> not using that transport (and thus going to SG_IO directly, which
> works).

I'm using /dev/hdc for burning in the k3b configuration screens, 
however that path may actually be defined.  I haven't quite "grok"ed 
all the details, but it works, and works with <10% of the cpu 
involved when burning.

To me, thats a roaring success :-)

-- 
Cheers, Gene
AMD K6-III@500mhz 320M
Athlon1600XP@1400mhz  512M
99.22% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attornies please note, additions to this message
by Gene Heskett are:
Copyright 2003 by Maurice Eugene Heskett, all rights reserved.

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Jens Axboe]

On Wed, Dec 17 2003, Gene Heskett wrote:
> On Wednesday 17 December 2003 11:49, Jens Axboe wrote:
> >On Wed, Dec 17 2003, Gene Heskett wrote:
> >> On Wednesday 17 December 2003 11:25, Thomas Voegtle wrote:
> >> >On Wed, 17 Dec 2003, Gene Heskett wrote:
> >> >> I take that it is attempting to scan all 8 addresses of the
> >> >> scsi bus even though its actually hitting the atapi stuff?  Or
> >> >> do I need an even fresher version of cdrecord? or libscg?
> >> >
> >> >Sorry, I shortend my output of cdrecord. With 2.6.0-test11 it
> >> > looks like this:
> >> >
> >> >Using libscg version 'schily-0.7'
> >> >scsibus0:
> >> >cdrecord: Warning: controller returns wrong size for CD
> >> > capabilities page. 0,0,0     0) 'CREATIVE' ' CD5233E        '
> >> > '2.05' Removable CD-ROM 0,1,0     1) 'PLEXTOR ' 'CD-R  
> >> > PX-W1610A' '1.04' Removable CD-ROM 0,2,0     2) *
> >> >        0,3,0     3) *
> >> >        0,4,0     4) *
> >> >        0,5,0     5) *
> >> >        0,6,0     6) *
> >> >        0,7,0     7) *
> >>
> >> I see.  I also don't see the warning you are getting, and this may
> >> be the reason you can't burn.  I also do not have the normal cdrom
> >> as device 0.  What happens if you swap the master/slave jumpers
> >> and put the recorder first?  It might be worth a try, and any
> >> changes in how it works would be a clue as to where the real
> >> stoppage is.
> >
> >The reason is that Thomas is using ATAPI which will go through
> >CDROM_SEND_PACKET (which is broken in -bk), while you are probably
> > not using that transport (and thus going to SG_IO directly, which
> > works).
> 
> I'm using /dev/hdc for burning in the k3b configuration screens, 
> however that path may actually be defined.  I haven't quite "grok"ed 
> all the details, but it works, and works with <10% of the cpu 
> involved when burning.
> 
> To me, thats a roaring success :-)

Excellent, glad to hear it :)

-- 
Jens Axboe

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[bill davidsen]

In article <200312171227.53913.gene.heskett@verizon.net>,
Gene Heskett  <gene.heskett@verizon.net> wrote:

| I'm using /dev/hdc for burning in the k3b configuration screens, 
| however that path may actually be defined.  I haven't quite "grok"ed 
| all the details, but it works, and works with <10% of the cpu 
| involved when burning.
| 
| To me, thats a roaring success :-)

I'm burning on a 2.4 kernel, using ide-scsi, and taking <1% CPU at 16x
burn, so you might check if your CD is set to use DMA, or if it's still
doing PIO. I use about 10% doing an audio burn, which does use PIO in
2.4 AFAIK.

You may be able to drop the CPU a good bit more.
-- 
bill davidsen <davidsen@tmr.com>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[bill davidsen]

In article <Pine.LNX.4.21.0312171604390.32339-100000@needs-no.brain.uni-freiburg.de>,
Thomas Voegtle  <thomas@voegtle-clan.de> wrote:
| 
| Hello,
|   
| cdrecord -dev=ATAPI -scanbus  with 2.6.0-test11-bk10 and bk13 shows this:
|   
| scsibus0:
|         0,0,0     0) '' '' '' NON CCS Disk
|         0,1,0     1) '' '' '' NON CCS Disk
| 
|   
| but this works well with 2.6.0-test11.
| =>
|   
|         0,0,0     0) 'CREATIVE' ' CD5233E        ' '2.05' Removable CD-ROM
|         0,1,0     1) 'PLEXTOR ' 'CD-R   PX-W1610A' '1.04' Removable CD-ROM
|   
| SuSE 9.0
|   
| please cc me, I'm not subscribed

I see the same results, also using wli-2. I'll try mm1 as soon as it
completes compilation.

I assume that this breakage is unintentional, but it's not clear if it's
a bug in the kernel or cdrecord not coping with something actually fixed
in the kernel.

I may get a chance to look at it tomorrow, I don't think I have a stock
test11 kernel left :-(
-- 
bill davidsen <davidsen@tmr.com>
  CTO, TMR Associates, Inc
Doing interesting things with little computers since 1979.

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Jens Axboe]

On Wed, Dec 17 2003, Thomas Voegtle wrote:
> 
> Hello,
>   
> cdrecord -dev=ATAPI -scanbus  with 2.6.0-test11-bk10 and bk13 shows this:
>   
> scsibus0:
>         0,0,0     0) '' '' '' NON CCS Disk
>         0,1,0     1) '' '' '' NON CCS Disk
> 
>   
> but this works well with 2.6.0-test11.
> =>
>   
>         0,0,0     0) 'CREATIVE' ' CD5233E        ' '2.05' Removable CD-ROM
>         0,1,0     1) 'PLEXTOR ' 'CD-R   PX-W1610A' '1.04' Removable CD-ROM
>   
> SuSE 9.0

Apply this to test11-bkLATEST

===== drivers/block/scsi_ioctl.c 1.38 vs edited =====
--- 1.38/drivers/block/scsi_ioctl.c	Thu Dec 11 18:55:17 2003
+++ edited/drivers/block/scsi_ioctl.c	Wed Dec 17 13:49:40 2003
@@ -150,13 +150,10 @@
 	struct request *rq;
 	struct bio *bio;
 	char sense[SCSI_SENSE_BUFFERSIZE];
-	unsigned char cdb[BLK_MAX_CDB];
 	void *buffer;
 
 	if (hdr->interface_id != 'S')
 		return -EINVAL;
-	if (hdr->cmd_len > sizeof(rq->cmd))
-		return -EINVAL;
 
 	/*
 	 * we'll do that later
@@ -167,9 +164,6 @@
 	if (hdr->dxfer_len > (q->max_sectors << 9))
 		return -EIO;
 
-	if (copy_from_user(cdb, hdr->cmdp, hdr->cmd_len))
-		return -EFAULT;
-
 	reading = writing = 0;
 	buffer = NULL;
 	bio = NULL;
@@ -220,7 +214,7 @@
 	 * fill in request structure
 	 */
 	rq->cmd_len = hdr->cmd_len;
-	memcpy(rq->cmd, cdb, hdr->cmd_len);
+	memcpy(rq->cmd, hdr->cmdp, hdr->cmd_len);
 	if (sizeof(rq->cmd) != hdr->cmd_len)
 		memset(rq->cmd + hdr->cmd_len, 0, sizeof(rq->cmd) - hdr->cmd_len);
 
@@ -436,12 +430,23 @@
 			break;
 		case SG_IO: {
 			struct sg_io_hdr hdr;
+			unsigned char cdb[BLK_MAX_CDB], *old_cdb;
 
-			if (copy_from_user(&hdr, (struct sg_io_hdr *) arg, sizeof(hdr))) {
-				err = -EFAULT;
+			err = -EFAULT;
+			if (copy_from_user(&hdr, (struct sg_io_hdr *) arg, sizeof(hdr)))
 				break;
-			}
+			err = -EINVAL;
+			if (hdr.cmd_len > sizeof(rq->cmd))
+				break;
+			err = -EFAULT;
+			if (copy_from_user(cdb, hdr.cmdp, hdr.cmd_len))
+				break;
+
+			old_cdb = hdr.cmdp;
+			hdr.cmdp = cdb;
 			err = sg_io(q, bdev, &hdr);
+
+			hdr.cmdp = old_cdb;
 			if (copy_to_user((struct sg_io_hdr *) arg, &hdr, sizeof(hdr)))
 				err = -EFAULT;
 			break;

-- 
Jens Axboe

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Thomas Voegtle]

On Wed, 17 Dec 2003, Jens Axboe wrote:

> 
> Apply this to test11-bkLATEST
> 
> ===== drivers/block/scsi_ioctl.c 1.38 vs edited =====
> --- 1.38/drivers/block/scsi_ioctl.c	Thu Dec 11 18:55:17 2003

Yes, this fixes the problem.
I have now 2.6.0-test11-bk13 + patch.

Thank you,

Thomas

-- 
 Thomas Vögtle    email: thomas@voegtle-clan.de
 ----- http://www.voegtle-clan.de/thomas ------

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Martin Schlemmer]

[-- Attachment #1: Type: text/plain, Size: 405 bytes --]

On Wed, 2003-12-17 at 19:35, Thomas Voegtle wrote:
> On Wed, 17 Dec 2003, Jens Axboe wrote:
> 
> > 
> > Apply this to test11-bkLATEST
> > 
> > ===== drivers/block/scsi_ioctl.c 1.38 vs edited =====
> > --- 1.38/drivers/block/scsi_ioctl.c	Thu Dec 11 18:55:17 2003
> 
> Yes, this fixes the problem.
> I have now 2.6.0-test11-bk13 + patch.
> 

Same this side, thanks.


-- 
Martin Schlemmer

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: no atapi cdrecord burning with 2.6.0-test11-bk10 / bk13

[Jens Axboe]

On Wed, Dec 17 2003, Thomas Voegtle wrote:
> On Wed, 17 Dec 2003, Jens Axboe wrote:
> 
> > 
> > Apply this to test11-bkLATEST
> > 
> > ===== drivers/block/scsi_ioctl.c 1.38 vs edited =====
> > --- 1.38/drivers/block/scsi_ioctl.c	Thu Dec 11 18:55:17 2003
> 
> Yes, this fixes the problem.
> I have now 2.6.0-test11-bk13 + patch.

Great, thanks for testing. Patch is in Linus' in-queue.

-- 
Jens Axboe