xmms segfaulting on 2.4.18 and 2.4.19-pre2-ac2 + oops

Hi !

I encounter problems with xmms segfaulting on two different kernels. The first one is 2.4.18-xfs-preempt-lockbreak-bttv. I use alsa drivers (0.90.0b12) for my SB PCI 128 card. The second one is 2.4.19-pre2-ac2 with shawn patch (xfs + rmap 12g) and preempt patch.

Here are the oops for the 2.4.18 :

ksymoops 2.4.4 on i686 2.4.18-xfs-preempt-lock-bttv. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.18-xfs-preempt-lock-bttv/ (default) -m /boot/System.map-2.4.18-xfs-preempt-lock-bttv (specified)

Unable to handle kernel paging request at virtual address d8d5c000
d8d56730
*pde = 17c45067
Oops: 0000
CPU:    0
EIP:    0010:[3c59x:__insmod_3c59x_S.bss_L40+828820/101914768]    Not tainted
EFLAGS: 00210202
eax: 00000001   ebx: 00000003   ecx: ffff389f   edx: fffff888
esi: fffff888   edi: d8d59fa6   ebp: d8d5bffe   esp: c445fe90
ds: 0018   es: 0018   ss: 0018
Process xmms (pid: 2679, stackpage=c445f000)
Stack: d8d56840 d8d56730 cae1c710 cae1c6f0 00000000 00000004 00000004 00000001 
       f871f871 00000c29 000003ec 00000400 cae1c680 c551b3c0 d8d56a8a cae1c680 
       c7a8b500 c7a8b480 00000400 000003ec cae1c680 00000400 cae1c680 00000400 
Call Trace: [3c59x:__insmod_3c59x_S.bss_L40+829092/101914496] [3c59x:__insmod_3c59x_S.bss_L40+828820/101914768] [3c59x:__insmod_3c59x_S.bss_L40+829678/101913910] [3c59x:__insmod_3c59x_S.bss_L40+817016/101926572] [3c59x:__insmod_3c59x_S.bss_L40+802037/101941551] 
Code: 8b 75 00 e9 87 00 00 00 8b 75 00 81 f6 00 80 00 00 eb 7c 8b 

Using defaults from ksymoops -t elf32-i386 -a i386

Code; 00000000 Before first symbol
00000000 <_EIP>:
Code; 00000000 Before first symbol
0: 8b 75 00 mov 0x0(%ebp),%esi
Code; 00000003 Before first symbol
3: e9 87 00 00 00 jmp 8f <_EIP+0x8f> 0000008f Before first symbol
Code; 00000008 Before first symbol
8: 8b 75 00 mov 0x0(%ebp),%esi
Code; 0000000b Before first symbol
b: 81 f6 00 80 00 00 xor $0x8000,%esi
Code; 00000011 Before first symbol
11: eb 7c jmp 8f <_EIP+0x8f> 0000008f Before first symbol
Code; 00000013 Before first symbol
13: 8b 00 mov (%eax),%eax

I don't have any problem with the network card after the oops, however I can't use the sound card any more (already used). xmms wasn't using the network. I didn't do a ksymoops on the another oops but it was located on 3c59x too.

3c59x: Donald Becker and others. www.scyld.com/network/vortex.html 00:0d.0: 3Com PCI 3c905 Boomerang 100baseTx at 0xb000. Vers LK1.1.16

With 2.4.17, I didn't have this problem.

I don't think its the 3com's driver fault, please run the decoded oops right after xmms crashes and make sure the map file specified is the correct one. Check out ksymoops(8) for perhaps more help.

Zwane

might be worth testing without preempt to rule that out...

OoO Lors de la soirée naissante du mercredi 06 mars 2002, vers 17:14, Arjan van de Ven <[email protected]> disait:

I will try to catch the next oops to its correctness then I will try without preempt.

panic("huh?\n");
	2.2.16 /usr/src/linux/arch/i386/kernel/smp.c

The oops is not in 3c59x. You are letting klogd convert the oops and klogd has been broken for years. <rant>Why do distributors insist on shipping such broken code?</rant>. Always run klogd with the -x flag to keep its sticky fingers off the oops then you can get clean data for ksymoops to decode.

OoO En cette soirée bien amorcée du mercredi 06 mars 2002, vers 22:26, Keith Owens <[email protected]> disait:

You are right, I have used the output of dmesg instead and here is what I get :

ksymoops 2.4.4 on i686 2.4.19-pre2-ac2-xfs-shawn9-preempt. Options used
-v /usr/src/linux/vmlinux (specified)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.19-pre2-ac2-xfs-shawn9-preempt/ (default) -m /usr/src/linux/System.map (default)

Reading Oops report from the terminal

Unable to handle kernel paging request at virtual address d8d7e000
d91a3730
*pde = 17b1f067
Oops: 0000
CPU:    0
EIP:    0010:[<d91a3730>]    Not tainted

Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00210202
eax: 00000001 ebx: 00000003 ecx: 000007ff edx: 00000000
esi: 00000000 edi: d8d7bfa6 ebp: d8d7dffe esp: c877de90
ds: 0018 es: 0018 ss: 0018
Process xmms (pid: 3910, stackpage=c877d000)

Stack: d91a3840 d91a3730 cb4a4150 cb4a4130 00000000 00000004 00000004 00000001 
       00000000 00000c29 000003ec 00000400 cb4a40c0 cd94cec0 d91a3a8a cb4a40c0 
       c4dc8340 c4dc8380 00000400 000003ec cb4a40c0 00000400 cb4a40c0 00000400 
Call Trace: [<d91a3840>] [<d91a3730>] [<d91a3a8a>] [<d91a0914>] [<d919ce91>] 
   [<d919d013>] [<d919ed23>] [<c0134ad5>] [<c0106f2b>] 
Code: 8b 75 00 e9 87 00 00 00 8b 75 00 81 f6 00 80 00 00 eb 7c 8b 

Trace; d91a3840 <[snd-pcm-oss]resample_shrink+290/390> Trace; d91a3730 <[snd-pcm-oss]resample_shrink+180/390> Trace; d91a3a8a <[snd-pcm-oss]rate_transfer+2a/40> Trace; d91a0914 <[snd-pcm-oss]snd_pcm_plug_write_transfer+94/d0> Trace; d919ce91 <[snd-pcm-oss]snd_pcm_oss_write2+81/d0> Trace; d919d013 <[snd-pcm-oss]snd_pcm_oss_write1+133/160> Trace; d919ed23 <[snd-pcm-oss]snd_pcm_oss_write+33/50> Trace; c0134ad5 <sys_write+95/120>
Trace; c0106f2b <system_call+33/38>
Code; d91a3730 <[snd-pcm-oss]resample_shrink+180/390> 00000000 <_EIP>:
Code; d91a3730 <[snd-pcm-oss]resample_shrink+180/390> <=====
0: 8b 75 00 mov 0x0(%ebp),%esi <=====
Code; d91a3733 <[snd-pcm-oss]resample_shrink+183/390>
3: e9 87 00 00 00 jmp 8f <_EIP+0x8f> d91a37bf <[snd-pcm-oss]resample_shrink+20f/390>
Code; d91a3738 <[snd-pcm-oss]resample_shrink+188/390>
8: 8b 75 00 mov 0x0(%ebp),%esi
Code; d91a373b <[snd-pcm-oss]resample_shrink+18b/390>
b: 81 f6 00 80 00 00 xor $0x8000,%esi
Code; d91a3741 <[snd-pcm-oss]resample_shrink+191/390>
11: eb 7c jmp 8f <_EIP+0x8f> d91a37bf <[snd-pcm-oss]resample_shrink+20f/390>
Code; d91a3743 <[snd-pcm-oss]resample_shrink+193/390>
13: 8b 00 mov (%eax),%eax

So, a problem with Alsa ? I use Alsa 0.9.0b12, compiled with no option (plain ./configure).

snd-pcm-oss            36192   1 (autoclean)
snd-mixer-oss           8912   0 (autoclean) [snd-pcm-oss]
joydev                  5808   1 (autoclean)
ns558                   1360   0 (autoclean) (unused)
analog                  7536   0 (autoclean) (unused)
input                   3200   0 (autoclean) [joydev analog]
gameport                1376   0 (autoclean) [ns558 analog]
tuner                   8512   1 (autoclean)
tvaudio                11232   0 (autoclean) (unused)
bttv                   65584   0 (autoclean)
i2c-algo-bit            6912   1 (autoclean) [bttv]
i2c-core               12704   0 (autoclean) [tuner tvaudio bttv i2c-algo-bit]
videodev                4992   2 (autoclean) [bttv]
usb-uhci               22144   0 (unused)
printer                 5632   0
usbcore                58016   0 [usb-uhci printer]
snd-seq-midi            3200   0 (unused)
snd-seq-midi-event      2976   0 [snd-seq-midi]
snd-seq                36304   0 [snd-seq-midi snd-seq-midi-event]
snd-ens1370             8352   1 (autoclean)
snd-pcm                49184   0 (autoclean) [snd-pcm-oss snd-ens1370]
snd-timer              10096   0 (autoclean) [snd-seq snd-pcm]
snd-rawmidi            12256   0 (autoclean) [snd-seq-midi snd-ens1370]
snd-seq-device          3856   0 (autoclean) [snd-seq-midi snd-seq snd-rawmidi]
snd-ak4531-codec        4896   0 (autoclean) [snd-ens1370]
snd                    24992   0 [snd-pcm-oss snd-mixer-oss snd-seq-midi snd-seq-midi-event snd-seq snd-ens1370 snd-pcm snd-timer snd-rawmidi snd-seq-device snd-ak4531-codec]
soundcore               3536   6 [snd]
ide-scsi                7840   0 (autoclean)
sr_mod                 13696   0 (autoclean)
scsi_mod               64048   2 (autoclean) [ide-scsi sr_mod]
cdrom                  27264   0 (autoclean) [sr_mod]
smbfs                  32848   1 (autoclean)
nfsd                   67648   4 (autoclean)
lockd                  48016   1 (autoclean) [nfsd]
sunrpc                 63264   1 (autoclean) [nfsd lockd]
3c59x                  25120   1 (autoclean)
rtc                     6496   0 (autoclean)

I have updated xmms to 1.2.7 too and I use the OSS plugin. I have crossposted to alsa mailing lists and it seems that it is a known problem. Feel free to drop lkml from the followup.

MY MOM IS NOT DATING JERRY SEINFELD
MY MOM IS NOT DATING JERRY SEINFELD
MY MOM IS NOT DATING JERRY SEINFELD
-+- Bart Simpson on chalkboard in episode AABF06

_______________________________________________
Alsa-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/alsa-devel