Bug? cdrecord hangs and uses up to 100% cpu

View: New views
7 Messages — Rating Filter:   Alert me  

Bug? cdrecord hangs and uses up to 100% cpu

by Bugzilla from trueg@k3b.org :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Hi Jörg, hi list,

I think I found a bug or at least a problem. This is the command line using
cdrecord 2.01.01a57:

cdrecord -v gracetime=2 dev=/dev/hda speed=40 -sao -dummy driveropts=burnfree
textfile=/tmp/Sn6771.dat -multi -useinfo -audio -pad -shorttrack /tmp/kde-
kde4dev/k3b_audio_0_01.wav

The problem is the textfile which is empty. Without it burning works
flawlessly. Now obviously using an empty textfile is bogus but it should not
make cdrecord go crazy, right?

Hope this helps,
Sebastian



# Cdrecord-Inf-File written by K3b 1.61.0, Do. Mrz 26 10:22:31 2009
#
ISRC=
MCN=
Albumperformer= ''
Albumtitle= ''
Performer= ''
Songwriter= ''
Composer= ''
Arranger= ''
Message= ''
Tracktitle= ''
Tracknumber= 1
Trackstart= 0
# Tracklength: 03:28:35
Tracklength= 15635, 0
Pre-emphasis= no
Channels= 2
Copy_permitted= yes
Endianess= little
Index= 0
Index0= -1

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

Re: Bug? cdrecord hangs and uses up to 100% cpu

by Joerg Schilling-3 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Sebastian Trüg <trueg@...> wrote:

> I think I found a bug or at least a problem. This is the command line using
> cdrecord 2.01.01a57:
>
> cdrecord -v gracetime=2 dev=/dev/hda speed=40 -sao -dummy driveropts=burnfree
> textfile=/tmp/Sn6771.dat -multi -useinfo -audio -pad -shorttrack /tmp/kde-
> kde4dev/k3b_audio_0_01.wav
>
> The problem is the textfile which is empty. Without it burning works
> flawlessly. Now obviously using an empty textfile is bogus but it should not
> make cdrecord go crazy, right?

I cannot reproduce a hang with an empty textfile. Could you call:

gcore <pid of cdrecord>
and use the debugger or

run
pstack  <pid of cdrecord>

to find where the hang is?

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

Re: Bug? cdrecord hangs and uses up to 100% cpu

by Bugzilla from trueg@k3b.org :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Thursday 26 March 2009 12:19:12 Joerg Schilling wrote:

> Sebastian Trüg <trueg@...> wrote:
> > I think I found a bug or at least a problem. This is the command line
> > using cdrecord 2.01.01a57:
> >
> > cdrecord -v gracetime=2 dev=/dev/hda speed=40 -sao -dummy
> > driveropts=burnfree textfile=/tmp/Sn6771.dat -multi -useinfo -audio -pad
> > -shorttrack /tmp/kde- kde4dev/k3b_audio_0_01.wav
> >
> > The problem is the textfile which is empty. Without it burning works
> > flawlessly. Now obviously using an empty textfile is bogus but it should
> > not make cdrecord go crazy, right?
>
> I cannot reproduce a hang with an empty textfile. Could you call:
>
> gcore <pid of cdrecord>
> and use the debugger or
>
> run
> pstack  <pid of cdrecord>
>
> to find where the hang is?

would love to but I dont have any of those commands. pstack failed to compile
here on amd64 and gcore is not available at all... :(
Any other output I may produce? I could compile cdrecord with debugging
symbols...
_______________________________________________
Cdrecord-developers mailing list
Cdrecord-developers@...
https://lists.berlios.de/mailman/listinfo/cdrecord-developers

Re: Bug? cdrecord hangs and uses up to 100% cpu

by Joerg Schilling-3 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Sebastian Trüg <trueg@...> wrote:

> > to find where the hang is?
>
> would love to but I dont have any of those commands. pstack failed to compile
> here on amd64 and gcore is not available at all... :(
> Any other output I may produce? I could compile cdrecord with debugging
> symbols...

if you have no working dbx (dbx prints a readable stack trace even for stripped
binaries), you would need to compile for gdb....see README.compile


Then you could e.g. call

kill -ILL <cdrecord pid>

and then call

gdb cdrecord core
where

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

Re: Bug? cdrecord hangs and uses up to 100% cpu

by Bugzilla from trueg@k3b.org :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Thursday 26 March 2009 13:17:31 Joerg Schilling wrote:

> Sebastian Trüg <trueg@...> wrote:
> > > to find where the hang is?
> >
> > would love to but I dont have any of those commands. pstack failed to
> > compile here on amd64 and gcore is not available at all... :(
> > Any other output I may produce? I could compile cdrecord with debugging
> > symbols...
>
> if you have no working dbx (dbx prints a readable stack trace even for
> stripped binaries), you would need to compile for gdb....see README.compile
>
>
> Then you could e.g. call
>
> kill -ILL <cdrecord pid>
>
> and then call
>
> gdb cdrecord core
> where
please find attached the backtrace. It is short but at least it shows the
area...

movebytes (fromv=0x1e49270, tov=0x1e5c000, cnt=0) at movebytes.c:42
42              if ((n = cnt) <= 0)
(gdb) bt
#0  movebytes (fromv=0x1e49270, tov=0x1e5c000, cnt=0) at movebytes.c:42
#1  0x00000000004104b0 in write_cdtext (scgp=0x1e49290, dp=0x1e5a8c0, startsec=-11933) at cdtext.c:480
#2  0x000000000041884c in write_leadin_mmc (scgp=0x1e49290, dp=0x1e5a8c0, trackp=0x7fff10949fd0) at drv_mmc.c:1847
#3  0x0000000000409f0e in main (ac=<value optimized out>, av=<value optimized out>) at cdrecord.c:1280

gdb) step
125     }
(gdb)
write_cdtext (scgp=0x1e49290, dp=0x1e5a8c0, startsec=-11933) at cdtext.c:479
479                     for (idx = 0; (idx + textlen) <= maxdma; idx += textlen)
(gdb)
480                             movebytes(textsub, &bp[idx], textlen);
(gdb)
movebytes (fromv=0x1e49270, tov=0x1e5c000, cnt=0) at movebytes.c:36
36              register char           *to     = tov;
(gdb)
42              if ((n = cnt) <= 0)
(gdb)
125     }
(gdb)
write_cdtext (scgp=0x1e49290, dp=0x1e5a8c0, startsec=-11933) at cdtext.c:479
479                     for (idx = 0; (idx + textlen) <= maxdma; idx += textlen)
(gdb)
480                             movebytes(textsub, &bp[idx], textlen);
(gdb)
movebytes (fromv=0x1e49270, tov=0x1e5c000, cnt=0) at movebytes.c:36
36              register char           *to     = tov;
(gdb)
42              if ((n = cnt) <= 0)
(gdb)
125     }
(gdb)
write_cdtext (scgp=0x1e49290, dp=0x1e5a8c0, startsec=-11933) at cdtext.c:479
479                     for (idx = 0; (idx + textlen) <= maxdma; idx += textlen)
(gdb)
480                             movebytes(textsub, &bp[idx], textlen);
(gdb)
movebytes (fromv=0x1e49270, tov=0x1e5c000, cnt=0) at movebytes.c:36
36              register char           *to     = tov;
(gdb)
42              if ((n = cnt) <= 0)
(gdb)
125     }

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

Re: Bug? cdrecord hangs and uses up to 100% cpu

by Joerg Schilling-3 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

Sebastian Trüg <trueg@...> wrote:

> please find attached the backtrace. It is short but at least it shows the
> area...

It seems that the hang happens lather than your first mail made be believe....

Please check the latest a59-pre2 in

ftp://ftp.berlios.de/pub/cdrecord/alpha/

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

Re: Bug? cdrecord hangs and uses up to 100% cpu

by Bugzilla from trueg@k3b.org :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Thursday 26 March 2009 15:49:13 Joerg Schilling wrote:

> Sebastian Trüg <trueg@...> wrote:
> > please find attached the backtrace. It is short but at least it shows the
> > area...
>
> It seems that the hang happens lather than your first mail made be
> believe....
>
> Please check the latest a59-pre2 in
>
> ftp://ftp.berlios.de/pub/cdrecord/alpha/
>
> Jörg

:D
nice solution:

/usr/local/bin/cdrecord: Empty CD-Text file.
/usr/local/bin/cdrecord: Cannot use '/tmp/Sn6771.dat' as CD-Text file.

Cheers,
Sebastian

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