* 2.4.20-pre2 NFS OOPS on sparc64
@ 2002-08-13 16:21 Meelis Roos
2002-08-13 17:27 ` David S. Miller
0 siblings, 1 reply; 5+ messages in thread
From: Meelis Roos @ 2002-08-13 16:21 UTC (permalink / raw)
To: linux-kernel
2 oopses from stock 2.4.20-pre2 during NFS startup 9mountd etc killed as
a result). Looks like a bad use of bitops inside sunrpc. egcs64 compiler
from debian.
ksymoops 2.4.6 on sparc64 2.4.20-pre2. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.4.20-pre2/ (default)
-m /boot/System.map-2.4.20-pre2 (default)
Warning: You did not tell me where to find symbol information. I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc. ksymoops -h explains the options.
Aug 13 19:08:24 janku kernel: ac97_codec: AC97 codec, id: 0x0000:0x0000 (Unknown)
Aug 13 19:08:29 janku kernel: \|/ ____ \|/
Aug 13 19:08:29 janku kernel: "@'/ .. \`@"
Aug 13 19:08:29 janku kernel: /_| \__/ |_\
Aug 13 19:08:29 janku kernel: \__U_/
Aug 13 19:08:29 janku kernel: rpc.nfsd(545): Kernel does fpu/atomic unaligned load/store.
Aug 13 19:08:29 janku kernel: TSTATE: 0000004480009602 TPC: 00000000005bfae8 TNPC: 00000000005bfaec Y: 07000000 Not tainted
Using defaults from ksymoops -t elf32-sparc -a sparc
Aug 13 19:08:29 janku kernel: g0: 0000000000000011 g1: 0000000000000080 g2: 0000000000000007 g3: 0000000000000000
Aug 13 19:08:29 janku kernel: g4: fffff80000000000 g5: 0000000000000080 g6: fffff8000e0fc000 g7: 0000000000000000
Aug 13 19:08:29 janku kernel: o0: 0000000000000000 o1: fffff8000fe630b4 o2: fffff8000078ea80 o3: 00000000005a5360
Aug 13 19:08:29 janku kernel: o4: 00000000005a56a0 o5: fffff8000fe63078 sp: fffff8000e0ff061 ret_pc: 00000000005a5748
Aug 13 19:08:29 janku kernel: l0: fffff8000fe28270 l1: fffff8000fe63078 l2: 00000000005c9000 l3: 0000000000000000
Aug 13 19:08:29 janku kernel: l4: 000000005a2cf071 l5: 0000000000000000 l6: fffff8000e0fc000 l7: 0000000000432f20
Aug 13 19:08:29 janku kernel: i0: 0000000000000000 i1: 000000003d592efc i2: 00000000006d0c00 i3: 000000000000001c
Aug 13 19:08:29 janku kernel: i4: 00000000005b43e4 i5: 0000000000000000 i6: fffff8000e0ff121 i7: 00000000005a6a40
Aug 13 19:08:29 janku kernel: Caller[00000000005a6a40]
Aug 13 19:08:29 janku kernel: Caller[00000000005a6cbc]
Aug 13 19:08:29 janku kernel: Caller[00000000005a6ec4]
Aug 13 19:08:29 janku kernel: Caller[000000000201614c]
Aug 13 19:08:29 janku kernel: Caller[0000000002016988]
Aug 13 19:08:29 janku kernel: Caller[000000000047f228]
Aug 13 19:08:29 janku kernel: Caller[0000000000433048]
Aug 13 19:08:29 janku kernel: Caller[0000000000410674]
Aug 13 19:08:29 janku kernel: Caller[0000000000010cec]
Aug 13 19:08:29 janku kernel: Instruction DUMP: 9089c005 12600006 8219c005 <c3f25007> 80a1c001 3267fffb ce5a4000 81c3e008 8143e00a
>>PC; 005bfae8 <__bitops_begin+28/40> <=====
>>o3; 005a5360 <svc_udp_recvfrom+0/340>
>>o4; 005a56a0 <svc_udp_sendto+0/60>
>>ret_pc; 005a5748 <svc_udp_init+48/60>
>>l2; 005c9000 <tvecs+488/d40>
>>l7; 00432f20 <sys32_nfsservctl+0/1e0>
>>i2; 006d0c00 <reserve.607+80/e0>
>>i4; 005b43e4 <__bzero+1d8/274>
>>i7; 005a6a40 <svc_setup_socket+a0/200>
Trace; 005a6a40 <svc_setup_socket+a0/200>
Trace; 005a6cbc <svc_create_socket+11c/180>
Trace; 005a6ec4 <svc_makesock+44/60>
Trace; 0201614c <[nfsd].text.start+8c/1a0>
Trace; 02016988 <[nfsd]handle_sys_nfsservctl+108/3c0>
Trace; 0047f228 <sys_nfsservctl+68/c0>
Trace; 00433048 <sys32_nfsservctl+128/1e0>
Trace; 00410674 <linux_sparc_syscall32+34/40>
Trace; 00010cec Before first symbol
Code; 005bfadc <__bitops_begin+1c/40>
00000000 <_PC>:
Code; 005bfadc <__bitops_begin+1c/40>
0: 90 89 c0 05 andcc %g7, %g5, %o0
Code; 005bfae0 <__bitops_begin+20/40>
4: 12 60 00 06 unknown
Code; 005bfae4 <__bitops_begin+24/40>
8: 82 19 c0 05 xor %g7, %g5, %g1
Code; 005bfae8 <__bitops_begin+28/40> <=====
c: c3 f2 50 07 unknown <=====
Code; 005bfaec <__bitops_begin+2c/40>
10: 80 a1 c0 01 cmp %g7, %g1
Code; 005bfaf0 <__bitops_begin+30/40>
14: 32 67 ff fb unknown
Code; 005bfaf4 <__bitops_begin+34/40>
18: ce 5a 40 00 unknown
Code; 005bfaf8 <__bitops_begin+38/40>
1c: 81 c3 e0 08 retl
Code; 005bfafc <__bitops_begin+3c/40>
20: 81 43 e0 0a unknown
Aug 13 19:08:32 janku kernel: \|/ ____ \|/
Aug 13 19:08:32 janku kernel: "@'/ .. \`@"
Aug 13 19:08:32 janku kernel: /_| \__/ |_\
Aug 13 19:08:32 janku kernel: \__U_/
Aug 13 19:08:32 janku kernel: amd(593): Kernel does fpu/atomic unaligned load/store.
Aug 13 19:08:32 janku kernel: TSTATE: 0000004411009600 TPC: 00000000005bfae8 TNPC: 00000000005bfaec Y: 07000000 Not tainted
Aug 13 19:08:32 janku kernel: g0: 0000000000000011 g1: 0000000000000080 g2: 0000000000000007 g3: 0000000000000000
Aug 13 19:08:32 janku kernel: g4: fffff80000000000 g5: 0000000000000080 g6: fffff8000d7e4000 g7: 0000000000000000
Aug 13 19:08:32 janku kernel: o0: 0000000000000000 o1: fffff8000fe63534 o2: fffff8000d8a19e0 o3: 00000000005a5360
Aug 13 19:08:32 janku kernel: o4: 00000000005a56a0 o5: fffff8000fe634f8 sp: fffff8000d7e6d21 ret_pc: 00000000005a5748
Aug 13 19:08:32 janku kernel: l0: fffff8000fe28270 l1: fffff8000fe634f8 l2: 00000000005c9000 l3: 0000000000000000
Aug 13 19:08:32 janku kernel: l4: 000000005a2cf071 l5: 0000000000049000 l6: 000000000003f400 l7: 0000000000049000
Aug 13 19:08:32 janku kernel: i0: 0000000000000000 i1: 000000003d592eff i2: 00000000006d0c00 i3: 000000000000001c
Aug 13 19:08:32 janku kernel: i4: 00000000005b43e4 i5: 0000000000000000 i6: fffff8000d7e6de1 i7: 00000000005a6a40
Aug 13 19:08:32 janku kernel: Caller[00000000005a6a40]
Aug 13 19:08:32 janku kernel: Caller[00000000005a6cbc]
Aug 13 19:08:32 janku kernel: Caller[00000000005a6ec4]
Aug 13 19:08:32 janku kernel: Caller[00000000004bfec8]
Aug 13 19:08:32 janku kernel: Caller[00000000004b1c84]
Aug 13 19:08:32 janku kernel: Caller[000000000046c8e8]
Aug 13 19:08:32 janku kernel: Caller[000000000046cae4]
Aug 13 19:08:32 janku kernel: Caller[000000000048076c]
Aug 13 19:08:32 janku kernel: Caller[0000000000480aa8]
Aug 13 19:08:32 janku kernel: Caller[000000000042ffa8]
Aug 13 19:08:32 janku kernel: Caller[0000000000410674]
Aug 13 19:08:32 janku kernel: Caller[000000007004affc]
Aug 13 19:08:32 janku kernel: Instruction DUMP: 9089c005 12600006 8219c005 <c3f25007> 80a1c001 3267fffb ce5a4000 81c3e008 8143e00a
>>PC; 005bfae8 <__bitops_begin+28/40> <=====
>>o3; 005a5360 <svc_udp_recvfrom+0/340>
>>o4; 005a56a0 <svc_udp_sendto+0/60>
>>ret_pc; 005a5748 <svc_udp_init+48/60>
>>l2; 005c9000 <tvecs+488/d40>
>>i2; 006d0c00 <reserve.607+80/e0>
>>i4; 005b43e4 <__bzero+1d8/274>
>>i7; 005a6a40 <svc_setup_socket+a0/200>
Trace; 005a6a40 <svc_setup_socket+a0/200>
Trace; 005a6cbc <svc_create_socket+11c/180>
Trace; 005a6ec4 <svc_makesock+44/60>
Trace; 004bfec8 <lockd_up+a8/1a0>
Trace; 004b1c84 <nfs_read_super+a04/a20>
Trace; 0046c8e8 <get_sb_nodev+48/c0>
Trace; 0046cae4 <do_kern_mount+84/160>
Trace; 0048076c <do_add_mount+6c/1a0>
Trace; 00480aa8 <do_mount+168/1a0>
Trace; 0042ffa8 <sys32_mount+108/160>
Trace; 00410674 <linux_sparc_syscall32+34/40>
Trace; 7004affc <END_OF_CODE+6dfe9c89/????>
Code; 005bfadc <__bitops_begin+1c/40>
00000000 <_PC>:
Code; 005bfadc <__bitops_begin+1c/40>
0: 90 89 c0 05 andcc %g7, %g5, %o0
Code; 005bfae0 <__bitops_begin+20/40>
4: 12 60 00 06 unknown
Code; 005bfae4 <__bitops_begin+24/40>
8: 82 19 c0 05 xor %g7, %g5, %g1
Code; 005bfae8 <__bitops_begin+28/40> <=====
c: c3 f2 50 07 unknown <=====
Code; 005bfaec <__bitops_begin+2c/40>
10: 80 a1 c0 01 cmp %g7, %g1
Code; 005bfaf0 <__bitops_begin+30/40>
14: 32 67 ff fb unknown
Code; 005bfaf4 <__bitops_begin+34/40>
18: ce 5a 40 00 unknown
Code; 005bfaf8 <__bitops_begin+38/40>
1c: 81 c3 e0 08 retl
Code; 005bfafc <__bitops_begin+3c/40>
20: 81 43 e0 0a unknown
1 warning issued. Results may not be reliable.
--
Meelis Roos (mroos@linux.ee)
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 2.4.20-pre2 NFS OOPS on sparc64
2002-08-13 16:21 2.4.20-pre2 NFS OOPS on sparc64 Meelis Roos
@ 2002-08-13 17:27 ` David S. Miller
2002-08-13 22:27 ` Neil Brown
0 siblings, 1 reply; 5+ messages in thread
From: David S. Miller @ 2002-08-13 17:27 UTC (permalink / raw)
To: mroos; +Cc: linux-kernel, neilb
From: Meelis Roos <mroos@linux.ee>
Date: Tue, 13 Aug 2002 19:21:30 +0300 (EEST)
2 oopses from stock 2.4.20-pre2 during NFS startup 9mountd etc killed as
a result). Looks like a bad use of bitops inside sunrpc. egcs64 compiler
from debian.
Neil, sk_flags in struct svc_sock may not be an int, bitops require
"long".
--- include/linux/sunrpc/svcsock.h.~1~ Tue Aug 13 10:37:10 2002
+++ include/linux/sunrpc/svcsock.h Tue Aug 13 10:37:15 2002
@@ -22,7 +22,7 @@
struct svc_serv * sk_server; /* service for this socket */
unsigned char sk_inuse; /* use count */
- unsigned int sk_flags;
+ unsigned long sk_flags;
#define SK_BUSY 0 /* enqueued/receiving */
#define SK_CONN 1 /* conn pending */
#define SK_CLOSE 2 /* dead or dying */
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 2.4.20-pre2 NFS OOPS on sparc64
2002-08-13 17:27 ` David S. Miller
@ 2002-08-13 22:27 ` Neil Brown
2002-08-13 23:10 ` Alan Cox
2002-08-13 23:37 ` David S. Miller
0 siblings, 2 replies; 5+ messages in thread
From: Neil Brown @ 2002-08-13 22:27 UTC (permalink / raw)
To: David S. Miller; +Cc: linux-kernel
On Tuesday August 13, davem@redhat.com wrote:
> From: Meelis Roos <mroos@linux.ee>
> Date: Tue, 13 Aug 2002 19:21:30 +0300 (EEST)
>
> 2 oopses from stock 2.4.20-pre2 during NFS startup 9mountd etc killed as
> a result). Looks like a bad use of bitops inside sunrpc. egcs64 compiler
> from debian.
>
> Neil, sk_flags in struct svc_sock may not be an int, bitops require
> "long".
I knew that.... but obviously not at the right time. Thanks.
Now if only Linus has told me (like you did) instead of just making
the change himself in 2.5, I would have got it right in 2.4..
Anyway, I'll make sure it gets to Marcelo (if he hasn't picked it up
already) and will feel suitably chastised.
NeilBrown
>
> --- include/linux/sunrpc/svcsock.h.~1~ Tue Aug 13 10:37:10 2002
> +++ include/linux/sunrpc/svcsock.h Tue Aug 13 10:37:15 2002
> @@ -22,7 +22,7 @@
>
> struct svc_serv * sk_server; /* service for this socket */
> unsigned char sk_inuse; /* use count */
> - unsigned int sk_flags;
> + unsigned long sk_flags;
> #define SK_BUSY 0 /* enqueued/receiving */
> #define SK_CONN 1 /* conn pending */
> #define SK_CLOSE 2 /* dead or dying */
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 2.4.20-pre2 NFS OOPS on sparc64
2002-08-13 22:27 ` Neil Brown
@ 2002-08-13 23:10 ` Alan Cox
2002-08-13 23:37 ` David S. Miller
1 sibling, 0 replies; 5+ messages in thread
From: Alan Cox @ 2002-08-13 23:10 UTC (permalink / raw)
To: Neil Brown; +Cc: David S. Miller, linux-kernel
On Tue, 2002-08-13 at 23:27, Neil Brown wrote:
> On Tuesday August 13, davem@redhat.com wrote:
> > From: Meelis Roos <mroos@linux.ee>
> > Date: Tue, 13 Aug 2002 19:21:30 +0300 (EEST)
> >
> > 2 oopses from stock 2.4.20-pre2 during NFS startup 9mountd etc killed as
> > a result). Looks like a bad use of bitops inside sunrpc. egcs64 compiler
> > from debian.
> >
> > Neil, sk_flags in struct svc_sock may not be an int, bitops require
> > "long".
>
> I knew that.... but obviously not at the right time. Thanks.
>
> Now if only Linus has told me (like you did) instead of just making
> the change himself in 2.5, I would have got it right in 2.4.
May be my fault. I fixed it in 2.2 the first time it cropped up and
apparently forgot to tell you so it propogated
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: 2.4.20-pre2 NFS OOPS on sparc64
2002-08-13 22:27 ` Neil Brown
2002-08-13 23:10 ` Alan Cox
@ 2002-08-13 23:37 ` David S. Miller
1 sibling, 0 replies; 5+ messages in thread
From: David S. Miller @ 2002-08-13 23:37 UTC (permalink / raw)
To: neilb; +Cc: linux-kernel
From: Neil Brown <neilb@cse.unsw.edu.au>
Date: Wed, 14 Aug 2002 08:27:46 +1000
Now if only Linus has told me (like you did) instead of just making
the change himself in 2.5, I would have got it right in 2.4..
Non-long pointers sent to bitops get a warning in 2.5.x so it just
automatically shows up in the build logs there.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2002-08-13 23:47 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-08-13 16:21 2.4.20-pre2 NFS OOPS on sparc64 Meelis Roos
2002-08-13 17:27 ` David S. Miller
2002-08-13 22:27 ` Neil Brown
2002-08-13 23:10 ` Alan Cox
2002-08-13 23:37 ` David S. Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox