cdrecord and 2.6.28.10 linux kernel hangs

[Michael Shell]

  Jörg,

Sorry to report some bad news. Using cdrecord cdrtools-2.01.01a60pre2
(which reports itself as Cdrecord-ProDVD-ProBD-Clone 2.01.01a59), I
get a bad hang (I think the entire IDE bus hangs as no other applications
seem to be able to do disk I/O) doing a simple -scanbus. I have a CDROM
reader on a SCSI card and a PATA DVD writer as /dev/hdc.

Under kernel 2.6.20 (this is an older AMD 32-bit i586 system) everything
is just fine:

[root]: cdrecord -scanbus
Cdrecord-ProDVD-ProBD-Clone 2.01.01a59 (i586-pc-linux-gnu) Copyright (C) 1995-2009 Jörg Schilling
Linux sg driver version: 3.5.27
Using libscg version 'schily-0.9'.
scsibus0:
        0,0,0     0) *
        0,1,0     1) *
        0,2,0     2) 'PLEXTOR ' 'CD-ROM PX-32TS  ' '1.03' Removable CD-ROM
        0,3,0     3) *
        0,4,0     4) *
        0,5,0     5) *
        0,6,0     6) *
        0,7,0     7) *
scsibus1001:
        1001,0,0 100100) '_NEC    ' 'DVD_RW ND-4551A ' '1-09' Removable CD-ROM
        1001,1,0 100101) *
        1001,2,0 100102) *
        1001,3,0 100103) *
        1001,4,0 100104) *
        1001,5,0 100105) *
        1001,6,0 100106) *
        1001,7,0 100107) *
[root]:


However, under the very latest 2.6.28.10 kernel, things go very wrong.
The first scsibus scan prints, but the IDE bus seems to hang up hard
when cdrecord scans scsibus1000. An error message "Too many
CD/DVD/BD-recorder targets found" is reported before any devices are
reported.

The last thing reported is something like:

scsibus1000:
   1000,0,0  100000)*


and after that I can't kill cdrecord with ctl-C or even do a ps in
another shell as that process hangs too. :(


This behavior probably indicates a kernel problem because even if
cdrecord were at fault, the whole IDE bus should not hang like that.

Anyway, I thought I'd warn you about it in case more people
encounter it.


    Mike Shell

_______________________________________________
Cdrecord-support mailing list
Cdrecord-support@...
https://lists.berlios.de/mailman/listinfo/cdrecord-support

[Joerg Schilling-3]

Michael Shell <list1@...> wrote:

> Sorry to report some bad news. Using cdrecord cdrtools-2.01.01a60pre2
> (which reports itself as Cdrecord-ProDVD-ProBD-Clone 2.01.01a59), I
> get a bad hang (I think the entire IDE bus hangs as no other applications
> seem to be able to do disk I/O) doing a simple -scanbus. I have a CDROM
> reader on a SCSI card and a PATA DVD writer as /dev/hdc.
>
> Under kernel 2.6.20 (this is an older AMD 32-bit i586 system) everything
> is just fine:
>
> [root]: cdrecord -scanbus
> Cdrecord-ProDVD-ProBD-Clone 2.01.01a59 (i586-pc-linux-gnu) Copyright (C) 1995-2009 Jörg Schilling
> Linux sg driver version: 3.5.27
> Using libscg version 'schily-0.9'.
> scsibus0:
>         0,0,0     0) *
>         0,1,0     1) *
>         0,2,0     2) 'PLEXTOR ' 'CD-ROM PX-32TS  ' '1.03' Removable CD-ROM
>         0,3,0     3) *
>         0,4,0     4) *
>         0,5,0     5) *
>         0,6,0     6) *
>         0,7,0     7) *
> scsibus1001:
>         1001,0,0 100100) '_NEC    ' 'DVD_RW ND-4551A ' '1-09' Removable CD-ROM
>         1001,1,0 100101) *
>         1001,2,0 100102) *
>         1001,3,0 100103) *
>         1001,4,0 100104) *
>         1001,5,0 100105) *
>         1001,6,0 100106) *
>         1001,7,0 100107) *
> [root]:
>
>
> However, under the very latest 2.6.28.10 kernel, things go very wrong.
> The first scsibus scan prints, but the IDE bus seems to hang up hard
> when cdrecord scans scsibus1000. An error message "Too many
> CD/DVD/BD-recorder targets found" is reported before any devices are
> reported.

Thank you for the information!

You may try to look into the kernel messages and other Linux kernel related
resources to find indications for the problem and report it to the Linux kernel
people. Did you report it?

Jörg

--
 EMail:joerg@... (home) Jörg Schilling D-13353 Berlin
       js@...                (uni)  
       joerg.schilling@... (work) Blog: http://schily.blogspot.com/
 URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily
_______________________________________________
Cdrecord-support mailing list
Cdrecord-support@...
https://lists.berlios.de/mailman/listinfo/cdrecord-support

[Michael Shell]

On Fri, 08 May 2009 10:31:43 +0200
Joerg.Schilling@... (Joerg Schilling) wrote:

> You may try to look into the kernel messages and other Linux
> kernel related resources to find indications for the
> problem and report it to the Linux kernel people.
> Did you report it?


  Jörg,

There does not seem to be anything related in the log messages, but
this is expected given that the system cannot write to the disk
during the event. :(

I did not report it yet. I was hoping that someone else watching
the list can reproduce it on their machine. I do not know which
version of the kernel first showed the problem. It happened
somewhere between 2.6.20 and 2.6.28.10.

I note that in that 2.6.28.10 kernel configuration that
SCSI emulation support is now labeled as DEPRECATED and
it states:

"If both this SCSI emulation and native ATAPI support
 are compiled into the kernel, the native support will
 be used."

On both my kernels both these configuration options are selected,
but I do not pass any special kernel boot options (e.g., scsi
emulation setup) for the (ATA) DVD writer drive.


  Mike


_______________________________________________
Cdrecord-support mailing list
Cdrecord-support@...
https://lists.berlios.de/mailman/listinfo/cdrecord-support