Skip to contentSkip to conversation

kernel oops with .74 snapshot.

DISCUSSION Daniel Pittman 3 participants 4 messages Jul 7, 2003

Subsystems (3)
Networking IPv4/IPv6 Arm/actions semi architecture
lore
Share & tools
lore

Daniel Pittman

author
#0 lore raw

I got the following series of oops reports when booting a .74 snapshot. Following is information on the latest changeset in the CVS export server, and the reports.

Aside from that, the new IO scheduler is a huge improvement. I miss it already. :/

Daniel

RCS file: /home/cvs/linux-2.5/ChangeSet,v
Working file: ChangeSet
head: 1.11487

[...]

description:
BitKeeper to RCS/CVS export
----------------------------
revision 1.11487
date: 2003/07/06 20:23:55; author: torvalds; state: Exp; lines: +0 -0 Simplify and speed up mmap read-around handling

[...]

BKrev: 3f08855bJ-EPEKDcfen7tS_rGIDesg
=============================================================================

The reports were:

Jul  7 09:47:31 enki kernel: Unable to handle kernel paging request at virtual address 06dd7f9a
Jul  7 09:47:31 enki kernel:  printing eip:
Jul  7 09:47:31 enki kernel: c0134960
Jul  7 09:47:31 enki kernel: *pde = 00000000
Jul  7 09:47:31 enki kernel: Oops: 0000 [#1]
Jul  7 09:47:31 enki kernel: CPU:    0
Jul  7 09:47:31 enki kernel: EIP:    0060:[kfree+48/98]    Not tainted
Jul  7 09:47:31 enki kernel: EFLAGS: 00010006
Jul  7 09:47:31 enki kernel: EIP is at kfree+0x30/0x62
Jul  7 09:47:31 enki kernel: eax: 00a00000   ebx: dcc1a900   ecx: dbedbf80   edx: 06dd7f9a
Jul  7 09:47:31 enki kernel: esi: 00000100   edi: 00000206   ebp: dbdf1e80   esp: dbe4bf50
Jul  7 09:47:31 enki kernel: ds: 007b   es: 007b   ss: 0068
Jul  7 09:47:31 enki kernel: Process netstat (pid: 637, threadinfo=dbe4a000 task=dc634780)
Jul  7 09:47:31 enki kernel: Stack: 00000000 00000001 dcc1a900 dbedbf80 dc14a810 c015e8fe 00000100 dbedbfa0 
Jul  7 09:47:31 enki kernel:        dbedbf80 dff64ec0 dc14a810 c0146636 dc14a810 dbedbf80 dbedbf80 dc73a380 
Jul  7 09:47:31 enki kernel:        00000000 dbe4a000 c0144fd8 dbedbf80 dc73a380 dbedbf80 0805f038 0805f038 
Jul  7 09:47:31 enki kernel: Call Trace:
Jul  7 09:47:31 enki kernel:  [seq_release_private+37/72] seq_release_private+0x25/0x48
Jul  7 09:47:31 enki kernel:  [__fput+173/188] __fput+0xad/0xbc
Jul  7 09:47:31 enki kernel:  [filp_close+77/121] filp_close+0x4d/0x79
Jul  7 09:47:31 enki kernel:  [sys_close+80/95] sys_close+0x50/0x5f
Jul  7 09:47:31 enki kernel:  [syscall_call+7/11] syscall_call+0x7/0xb
Jul  7 09:47:31 enki kernel: 
Jul  7 09:47:31 enki kernel: Code: 8b 1a 8b 03 3b 43 04 73 19 89 74 83 10 83 03 01 57 9d 8b 5c 
Jul  7 09:52:31 enki kernel: Unable to handle kernel paging request at virtual address 06dd7f9a
Jul  7 09:52:31 enki kernel:  printing eip:
Jul  7 09:52:31 enki kernel: c0134960
Jul  7 09:52:31 enki kernel: *pde = 00000000
Jul  7 09:52:31 enki kernel: Oops: 0000 [#2]
Jul  7 09:52:31 enki kernel: CPU:    0
Jul  7 09:52:31 enki kernel: EIP:    0060:[kfree+48/98]    Not tainted
Jul  7 09:52:31 enki kernel: EFLAGS: 00010006
Jul  7 09:52:31 enki kernel: EIP is at kfree+0x30/0x62
Jul  7 09:52:31 enki kernel: eax: 00a00000   ebx: d1c9d680   ecx: cad0cc80   edx: 06dd7f9a
Jul  7 09:52:31 enki kernel: esi: 00000100   edi: 00000206   ebp: dbdf1e80   esp: cb897f50
Jul  7 09:52:31 enki kernel: ds: 007b   es: 007b   ss: 0068
Jul  7 09:52:31 enki kernel: Process netstat (pid: 1763, threadinfo=cb896000 task=c81fb280)
Jul  7 09:52:31 enki kernel: Stack: 00000000 00000004 d1c9d680 cad0cc80 dc14a810 c015e8fe 00000100 cad0cca0 
Jul  7 09:52:31 enki kernel:        cad0cc80 dff64ec0 dc14a810 c0146636 dc14a810 cad0cc80 cad0cc80 d4d92100 
Jul  7 09:52:31 enki kernel:        00000000 cb896000 c0144fd8 cad0cc80 d4d92100 cad0cc80 0805f038 0805f038 
Jul  7 09:52:31 enki kernel: Call Trace:
Jul  7 09:52:31 enki kernel:  [seq_release_private+37/72] seq_release_private+0x25/0x48
Jul  7 09:52:31 enki kernel:  [__fput+173/188] __fput+0xad/0xbc
Jul  7 09:52:31 enki kernel:  [filp_close+77/121] filp_close+0x4d/0x79
Jul  7 09:52:31 enki kernel:  [sys_close+80/95] sys_close+0x50/0x5f
Jul  7 09:52:31 enki kernel:  [syscall_call+7/11] syscall_call+0x7/0xb
Jul  7 09:52:31 enki kernel: 
Jul  7 09:52:31 enki kernel: Code: 8b 1a 8b 03 3b 43 04 73 19 89 74 83 10 83 03 01 57 9d 8b 5c 
Jul  7 09:57:32 enki kernel:  <1>Unable to handle kernel paging request at virtual address 06dd7f9a
Jul  7 09:57:32 enki kernel:  printing eip:
Jul  7 09:57:32 enki kernel: c0134960
Jul  7 09:57:32 enki kernel: *pde = 00000000
Jul  7 09:57:32 enki kernel: Oops: 0000 [#3]
Jul  7 09:57:32 enki kernel: CPU:    0
Jul  7 09:57:32 enki kernel: EIP:    0060:[kfree+48/98]    Not tainted
Jul  7 09:57:32 enki kernel: EFLAGS: 00010006
Jul  7 09:57:32 enki kernel: EIP is at kfree+0x30/0x62
Jul  7 09:57:32 enki kernel: eax: 00a00000   ebx: c7a38500   ecx: d92f2e80   edx: 06dd7f9a
Jul  7 09:57:32 enki kernel: esi: 00000100   edi: 00000206   ebp: dbdf1e80   esp: c9a63f50
Jul  7 09:57:32 enki kernel: ds: 007b   es: 007b   ss: 0068
Jul  7 09:57:32 enki kernel: Process netstat (pid: 1766, threadinfo=c9a62000 task=d92b4780)
Jul  7 09:57:32 enki kernel: Stack: 00000000 00000004 c7a38500 d92f2e80 dc14a810 c015e8fe 00000100 d92f2ea0 
Jul  7 09:57:32 enki kernel:        d92f2e80 dff64ec0 dc14a810 c0146636 dc14a810 d92f2e80 d92f2e80 d928ed00 
Jul  7 09:57:32 enki kernel:        00000000 c9a62000 c0144fd8 d92f2e80 d928ed00 d92f2e80 0805f038 0805f038 
Jul  7 09:57:32 enki kernel: Call Trace:
Jul  7 09:57:32 enki kernel:  [seq_release_private+37/72] seq_release_private+0x25/0x48
Jul  7 09:57:32 enki kernel:  [__fput+173/188] __fput+0xad/0xbc
Jul  7 09:57:32 enki kernel:  [filp_close+77/121] filp_close+0x4d/0x79
Jul  7 09:57:32 enki kernel:  [sys_close+80/95] sys_close+0x50/0x5f
Jul  7 09:57:32 enki kernel:  [syscall_call+7/11] syscall_call+0x7/0xb
Jul  7 09:57:32 enki kernel: 
Jul  7 09:57:32 enki kernel: Code: 8b 1a 8b 03 3b 43 04 73 19 89 74 83 10 83 03 01 57 9d 8b 5c 
Jul  7 10:00:02 enki kernel:  <4>process `dig' is using obsolete setsockopt SO_BSDCOMPAT
Jul  7 10:02:32 enki kernel: Unable to handle kernel NULL pointer dereference at virtual address 00000000
Jul  7 10:02:32 enki kernel:  printing eip:
Jul  7 10:02:32 enki kernel: c0134962
Jul  7 10:02:32 enki kernel: *pde = 00000000
Jul  7 10:02:32 enki kernel: Oops: 0000 [#4]
Jul  7 10:02:32 enki kernel: CPU:    0
Jul  7 10:02:32 enki kernel: EIP:    0060:[kfree+50/98]    Not tainted
Jul  7 10:02:32 enki kernel: EFLAGS: 00010006
Jul  7 10:02:32 enki kernel: EIP is at kfree+0x32/0x62
Jul  7 10:02:32 enki kernel: eax: 00a00000   ebx: 00000000   ecx: cc638c00   edx: 8f530080
Jul  7 10:02:32 enki kernel: esi: 00000100   edi: 00000206   ebp: dbdf1e80   esp: c5bd3f50
Jul  7 10:02:32 enki kernel: ds: 007b   es: 007b   ss: 0068
Jul  7 10:02:32 enki kernel: Process netstat (pid: 2462, threadinfo=c5bd2000 task=c8ddad80)
Jul  7 10:02:32 enki kernel: Stack: 00000000 00000004 c45bacc0 cc638c00 dc14a810 c015e8fe 00000100 cc638c20 
Jul  7 10:02:32 enki kernel:        cc638c00 dff64ec0 dc14a810 c0146636 dc14a810 cc638c00 cc638c00 cc4d5180 
Jul  7 10:02:32 enki kernel:        00000000 c5bd2000 c0144fd8 cc638c00 cc4d5180 cc638c00 0805f038 0805f038 
Jul  7 10:02:32 enki kernel: Call Trace:
Jul  7 10:02:32 enki kernel:  [seq_release_private+37/72] seq_release_private+0x25/0x48
Jul  7 10:02:32 enki kernel:  [__fput+173/188] __fput+0xad/0xbc
Jul  7 10:02:32 enki kernel:  [filp_close+77/121] filp_close+0x4d/0x79
Jul  7 10:02:32 enki kernel:  [sys_close+80/95] sys_close+0x50/0x5f
Jul  7 10:02:32 enki kernel:  [syscall_call+7/11] syscall_call+0x7/0xb
Jul  7 10:02:32 enki kernel: 
Jul  7 10:02:32 enki kernel: Code: 8b 03 3b 43 04 73 19 89 74 83 10 83 03 01 57 9d 8b 5c 24 08
Signature · 3 lines
Artists are people driven by a conflict between the desire to
communicate and the even stronger desire to hide.
        -- D. W. Winnicott
#1 lore raw
Quoted 4 lines · Daniel Pittman

I got the following series of oops reports when booting a .74 snapshot. Following is information on the latest changeset in the CVS export server, and the reports.

Just out of interest, does this fix it for you? It looks sane, but since David is off for the weekend, I don't want to apply it without some serious feedback that "yes, it fixes the problem".

Linus

Index: linux-2.5/net/ipv4/igmp.c
===================================================================
RCS file: /home/cvs/linux-2.5/net/ipv4/igmp.c,v
retrieving revision 1.29

net/ipv4/igmp.c +2 -2
diff -u -r1.29 igmp.c
@@ -2099,7 +2099,7 @@
 	struct in_device *in_dev;
 };
 
-#define	igmp_mc_seq_private(seq)	((struct igmp_mc_iter_state *)&seq->private)
+#define	igmp_mc_seq_private(seq)	((struct igmp_mc_iter_state *)(seq)->private)
 
 static inline struct ip_mc_list *igmp_mc_get_first(struct seq_file *seq)
 {

@@ -2254,7 +2254,7 @@
 	struct ip_mc_list *im;
 };
 
-#define igmp_mcf_seq_private(seq)	((struct igmp_mcf_iter_state *)&seq->private)
+#define igmp_mcf_seq_private(seq)	((struct igmp_mcf_iter_state *)(seq)->private)
 
 static inline struct ip_sf_list *igmp_mcf_get_first(struct seq_file *seq)
 {

Index: linux-2.5/net/ipv4/raw.c
===================================================================
RCS file: /home/cvs/linux-2.5/net/ipv4/raw.c,v
retrieving revision 1.32

net/ipv4/raw.c +1 -1
diff -u -r1.32 raw.c
@@ -687,7 +687,7 @@
 	int bucket;
 };
 
-#define raw_seq_private(seq) ((struct raw_iter_state *)&seq->private)
+#define raw_seq_private(seq) ((struct raw_iter_state *)(seq)->private)
 
 static struct sock *raw_get_first(struct seq_file *seq)
 {

Index: linux-2.5/net/ipv6/anycast.c
===================================================================
RCS file: /home/cvs/linux-2.5/net/ipv6/anycast.c,v retrieving revision 1.4

net/ipv6/anycast.c +1 -1
diff -u -r1.4 anycast.c
@@ -441,7 +441,7 @@
 	struct inet6_dev *idev;
 };
 
-#define ac6_seq_private(seq)	((struct ac6_iter_state *)&seq->private)
+#define ac6_seq_private(seq)	((struct ac6_iter_state *)(seq)->private)
 
 static inline struct ifacaddr6 *ac6_get_first(struct seq_file *seq)
 {

Index: linux-2.5/net/ipv6/ip6_flowlabel.c
===================================================================
RCS file: /home/cvs/linux-2.5/net/ipv6/ip6_flowlabel.c,v retrieving revision 1.5

net/ipv6/ip6_flowlabel.c +1 -1
diff -u -r1.5 ip6_flowlabel.c
@@ -559,7 +559,7 @@
 	int bucket;
 };
 
-#define ip6fl_seq_private(seq)	((struct ip6fl_iter_state *)&(seq)->private)
+#define ip6fl_seq_private(seq)	((struct ip6fl_iter_state *)(seq)->private)
 
 static struct ip6_flowlabel *ip6fl_get_first(struct seq_file *seq)
 {

Index: linux-2.5/net/ipv6/mcast.c
===================================================================
RCS file: /home/cvs/linux-2.5/net/ipv6/mcast.c,v
retrieving revision 1.25

net/ipv6/mcast.c +2 -2
diff -u -r1.25 mcast.c
@@ -2045,7 +2045,7 @@
 	struct inet6_dev *idev;
 };
 
-#define igmp6_mc_seq_private(seq)	((struct igmp6_mc_iter_state *)&seq->private)
+#define igmp6_mc_seq_private(seq)	((struct igmp6_mc_iter_state *)(seq)->private)
 
 static inline struct ifmcaddr6 *igmp6_mc_get_first(struct seq_file *seq)
 {

@@ -2185,7 +2185,7 @@
 	struct ifmcaddr6 *im;
 };
 
-#define igmp6_mcf_seq_private(seq)	((struct igmp6_mcf_iter_state *)&seq->private)
+#define igmp6_mcf_seq_private(seq)	((struct igmp6_mcf_iter_state *)(seq)->private)
 
 static inline struct ip6_sf_list *igmp6_mcf_get_first(struct seq_file *seq)
 {

Index: linux-2.5/net/ipv6/raw.c
===================================================================
RCS file: /home/cvs/linux-2.5/net/ipv6/raw.c,v
retrieving revision 1.31

net/ipv6/raw.c +1 -1
diff -u -r1.31 raw.c
@@ -913,7 +913,7 @@
 	int bucket;
 };
 
-#define raw6_seq_private(seq) ((struct raw6_iter_state *)&seq->private)
+#define raw6_seq_private(seq) ((struct raw6_iter_state *)(seq)->private)
 
 static struct sock *raw6_get_first(struct seq_file *seq)
 {
#2 lore raw

Em Sun, Jul 06, 2003 at 06:19:29PM -0700, Linus Torvalds escreveu:

Quoted 9 lines
Quoted 4 lines · Daniel Pittman

I got the following series of oops reports when booting a .74 snapshot. Following is information on the latest changeset in the CVS export server, and the reports.

Just out of interest, does this fix it for you? It looks sane, but since David is off for the weekend, I don't want to apply it without some serious feedback that "yes, it fixes the problem".

Yes, that was what Yoshfuji fixed (he introduced the bug) :-)

- Arnaldo

author
#3 lore raw
In reply to Linus Torvalds
Quoted 9 lines · Linus Torvalds
Quoted 4 lines · Daniel Pittman

I got the following series of oops reports when booting a .74 snapshot. Following is information on the latest changeset in the CVS export server, and the reports.

Just out of interest, does this fix it for you? It looks sane, but since David is off for the weekend, I don't want to apply it without some serious feedback that "yes, it fixes the problem".

Yes. With that applied, all the problems are gone.
Daniel

Signature · 5 lines
We wait and think that we are biding our time.
We keep silent, we believe in our strength.
We trust in propaganda of the wonders of tomorrow 
in the shadow of content we are victims in the end. 
        -- Covenant, _Theremin_