Skip to contentSkip to conversation

2.6.17-rc6-rt3

DISCUSSION Ingo Molnar 8 participants 20 messages Jun 10, 2006 – Jun 16, 2006

lore
Share & tools
lore
#0 lore raw

i have released the 2.6.17-rc6-rt3 tree, which can be downloaded from the usual place:

http://redhat.com/~mingo/realtime-preempt/

this is a fixes-only release: lots of fixes from Thomas Gleixner (for the softirq problem that caused those ping latency weirdnesses, for hrtimers and timers problems and for the RCU related bug that was causing instability and more), John Stultz, Jan Altenberg and Clark Williams. MIPS update from Manish Lachwani. Futex fix from Dinakar Guniguntala. It also includes the RT-scheduling SMP fix that could fix the scheduling problem reported by Darren Hart.

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

to build a 2.6.17-rc6-rt3 tree, the following patches should be applied:

http://kernel.org/pub/linux/kernel/v2.6/linux-2.6.16.tar.bz2
http://kernel.org/pub/linux/kernel/v2.6/testing/patch-2.6.17-rc6.bz2
http://redhat.com/~mingo/realtime-preempt/patch-2.6.17-rc6-rt3

Ingo

#1 lore raw

Hi,

Quoted 5 lines · Ingo Molnar

i have released the 2.6.17-rc6-rt3 tree, which can be downloaded from the usual place:

http://redhat.com/~mingo/realtime-preempt/

My system hangs on boot.

Here is bug http://www.stardust.webpages.pl/files/rt/2.6.17-rc6-rt3/bug1.jpg
Here is config http://www.stardust.webpages.pl/files/rt/2.6.17-rc6-rt3/rt-config

Regards,
Michal

Signature · 3 lines
Michal K. K. Piotrowski
LTG - Linux Testers Group
(http://www.stardust.webpages.pl/ltg/wiki/)
#5 lore raw
In reply to Ingo Molnar
Quoted 2 lines · Ingo Molnar

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

I still see two oddites.

top - 14:17:04 up 5 min, 8 users, load average: 0.37, 1.39, 0.78
Tasks: 148 total, 1 running, 147 sleeping, 0 stopped, 0 zombie
Cpu(s): 6.4% us, 3.0% sy, 0.0% ni, 90.6% id, 0.0% wa, 0.0% hi, 0.0% si

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND

 1281 root     -44  -5     0    0    0 S  2.0  0.0   0:00.57 IRQ 20
 6710 root      15   0  147m  15m 2196 S  2.0  1.6   0:11.85 X
 8144 root      15   0 29612  14m  10m S  2.0  1.5   0:05.09 kdesktop
 8087 root      15   0 27124  12m 8436 S  1.0  1.2   0:06.74 kxkb
 8135 root      15   0 24132 9124 6712 S  1.0  0.9   0:02.52 kaccess
 8139 root      15   0 27472  12m 8996 S  1.0  1.3   0:05.61 kwin
 8146 root      15   0 31028  14m  10m S  1.0  1.5   0:07.94 kicker
 8149 root      15   0 25380  10m 7948 S  1.0  1.0   0:02.30 klipper
 8154 root      15   0 29648  12m 8432 S  1.0  1.2   0:03.09 suseplugger
 8156 root      15   0 30912  14m  10m S  1.0  1.5   0:05.84 konsole
 8158 root      15   0 28072  13m 9844 S  1.0  1.4   0:07.30 kmix
 8159 root      15   0 30764  14m  10m S  1.0  1.5   0:05.99 konsole
 8165 root      15   0 30768  14m  10m S  1.0  1.5   0:06.26 konsole
 8170 root      15   0 35368  16m  12m S  1.0  1.7   0:04.47 konqueror
 8172 root      15   0 44876  10m 8100 S  1.0  1.1   0:03.16 knotify
 8297 root      16   0  2136 1076  788 R  1.0  0.1   0:02.51 top
    1 root      16   0   688  260  224 S  0.0  0.0   0:03.14 init

KDE twiddling it's thumbs takes 10% CPU, but didn't in rt29.

Fully repeatable oops when glibc's make check hits rt/tst-cpuclock1. This isn't a regression though, it's in rt29 too.

kernel BUG at :36841! <-- that's fully repeatable too invalid opcode: 0000 [#1]
PREEMPT SMP

Modules linked in: xt_pkttype ipt_LOG xt_limit snd_pcm_oss snd_mixer_oss snd_seq snd_seq_device edd tda9887 saa7134 prism54 ohci1394 ieee1394 ir_kbd_i2c bt878 snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer snd soundcore snd_page_alloc i2c_i801 ip6t_REJECT xt_tcpudp ipt_REJECT xt_state iptable_mangle iptable_nat ip_nat iptable_filter ip6table_mangle ip_conntrack nfnetlink ip_tables ip6table_filter ip6_tables x_tables tuner bttv video_buf firmware_class ir_common btcx_risc tveeprom sd_mod nls_iso8859_1 nls_cp437 nls_utf8
CPU:    1
EIP:    0060:[<b103cbaa>]    Not tainted VLI
EFLAGS: 00010202   (2.6.17-rc6-rt3-smp #169) 

EIP is at posix_cpu_timer_set+0x505/0x52e
eax: 00000282 ebx: c9380f6c ecx: ef8b0b90 edx: dff80e10
esi: 3b9aca00 edi: c9380ee4 ebp: c9380eac esp: c9380e54
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process ld-linux.so.2 (pid: 18428, threadinfo=c9380000 task=dff80e10 stack_left=3616 worst_left=-1)

Stack: b13f98ed 00000202 c9380edc c9380ee4 c9380e78 b10151dc 00000000 05f5e100 
       00000000 05f5e100 00000000 ef8b0b90 fffffffd 00000000 c9380ee4 c9380ea4 
       b10151dc 00000000 00000000 c9380f6c 00000000 c9380ee4 c9380f88 b103ce18 
Call Trace:
 [<b10044db>] show_stack_log_lvl+0xaa/0xd5 (32)
 [<b10046c8>] show_registers+0x1c2/0x28e (68)
 [<b10048d0>] die+0x13c/0x31d (60)
 [<b1004b3b>] do_trap+0x8a/0xdb (32)
 [<b1005589>] do_invalid_op+0xae/0xb8 (192)
 [<b1003f97>] error_code+0x4f/0x54 (148)
 [<b103ce18>] posix_cpu_nsleep+0xfd/0x23b (220)
 [<b103969f>] sys_clock_nanosleep+0xe7/0xee (44)
 [<b10033e4>] syscall_call+0x7/0xb (-4020)
Code: 68 00 00 00 00 c7 45 d8 fd ff ff ff e9 db fd ff ff c7 41 6c ff ff ff ff c7 45 d8 01 00 00 00 c7 45 e4 01 00 00 00 e9 e8 fb ff ff <0f> 0b e9 8f fb ff ff b8 80 7a 5a b1 e8 ff cc 3b 00 e9 81 fd ff 
EIP: [<b103cbaa>] posix_cpu_timer_set+0x505/0x52e SS:ESP 0068:c9380e54

<peek>
(gdb) list *posix_cpu_timer_set+0x505
0xb103cbaa is in posix_cpu_timer_set (posix-cpu-timers.c:724).
719 }
720
721 /*
722 * Disarm any old timer after extracting its expiry time.
723 */
724 BUG_ON(!irqs_disabled());
725

726             ret = 0;
727             spin_lock(&p->sighand->siglock);
728             old_expires = timer->it.cpu.expires;

(gdb) list *posix_cpu_nsleep+0xfd
0xb103ce18 is in posix_cpu_nsleep (posix-cpu-timers.c:1597).
1592 static struct itimerspec zero_it;
1593 struct itimerspec it = { .it_value = *rqtp,
1594 .it_interval = {} };
1595

1596                    spin_lock_irq(&timer.it_lock);
1597                    error = posix_cpu_timer_set(&timer, flags, &it, NULL);
1598                    if (error) {

1599 spin_unlock_irq(&timer.it_lock);
1600 return error;
1601 }
(gdb)
<nope, definitely not Kansas>

#3 lore raw

Hi Thomas,

Quoted 7 lines · Thomas Gleixner

Michal,

Quoted 1 line · Michal Piotrowski

My system hangs on boot.

It boots fine with your config here :(. Any chance to get a full output via serial console ?

Currently not. I'll buy serial cable.

Quoted 3 lines

tglx

Regards,
Michal

Signature · 3 lines
Michal K. K. Piotrowski
LTG - Linux Testers Group
(http://www.stardust.webpages.pl/ltg/wiki/)
#4 lore raw
Quoted 5 lines · Michal Piotrowski

Hi Thomas,

Quoted 2 lines · Thomas Gleixner

Michal,

Quoted 8 lines
Quoted 5 lines · Michal Piotrowski
Quoted 1 line

My system hangs on boot.

It boots fine with your config here :(. Any chance to get a full output via serial console ?

Currently not. I'll buy serial cable.

Everything is ok when I build a kernel with gcc 4.1 [michal@ltg01-fedora ~]$ gcc -v
Using built-in specs.
Target: i386-redhat-linux
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux
Thread model: posix
gcc version 4.1.1 20060525 (Red Hat 4.1.1-1)

Can you try build a kernel with gcc 3.4.6?

[michal@ltg01-fedora ~]$ gcc-3.4 -v
Reading specs from /usr/local/bin/../lib/gcc/i686-pc-linux-gnu/3.4.6/specs Configured with: ./configure --prefix=/usr/local/ --disable-nls --enable-shared --enable-languages=c --program-suffix=-3.4 Thread model: posix
gcc version 3.4.6

Regards,
Michal

Signature · 3 lines
Michal K. K. Piotrowski
LTG - Linux Testers Group
(http://www.stardust.webpages.pl/ltg/wiki/)
#6 lore raw
In reply to Mike Galbraith
Quoted 54 lines · Mike Galbraith
Quoted 2 lines · Ingo Molnar

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

Fully repeatable oops when glibc's make check hits rt/tst-cpuclock1. This isn't a regression though, it's in rt29 too.

kernel BUG at :36841! <-- that's fully repeatable too invalid opcode: 0000 [#1]
PREEMPT SMP

Modules linked in: xt_pkttype ipt_LOG xt_limit snd_pcm_oss snd_mixer_oss snd_seq snd_seq_device edd tda9887 saa7134 prism54 ohci1394 ieee1394 ir_kbd_i2c bt878 snd_intel8x0 snd_ac97_codec snd_ac97_bus snd_pcm snd_timer snd soundcore snd_page_alloc i2c_i801 ip6t_REJECT xt_tcpudp ipt_REJECT xt_state iptable_mangle iptable_nat ip_nat iptable_filter ip6table_mangle ip_conntrack nfnetlink ip_tables ip6table_filter ip6_tables x_tables tuner bttv video_buf firmware_class ir_common btcx_risc tveeprom sd_mod nls_iso8859_1 nls_cp437 nls_utf8
CPU:    1
EIP:    0060:[<b103cbaa>]    Not tainted VLI
EFLAGS: 00010202   (2.6.17-rc6-rt3-smp #169) 

EIP is at posix_cpu_timer_set+0x505/0x52e
eax: 00000282 ebx: c9380f6c ecx: ef8b0b90 edx: dff80e10
esi: 3b9aca00 edi: c9380ee4 ebp: c9380eac esp: c9380e54
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process ld-linux.so.2 (pid: 18428, threadinfo=c9380000 task=dff80e10 stack_left=3616 worst_left=-1)

Stack: b13f98ed 00000202 c9380edc c9380ee4 c9380e78 b10151dc 00000000 05f5e100 
       00000000 05f5e100 00000000 ef8b0b90 fffffffd 00000000 c9380ee4 c9380ea4 
       b10151dc 00000000 00000000 c9380f6c 00000000 c9380ee4 c9380f88 b103ce18 
Call Trace:
 [<b10044db>] show_stack_log_lvl+0xaa/0xd5 (32)
 [<b10046c8>] show_registers+0x1c2/0x28e (68)
 [<b10048d0>] die+0x13c/0x31d (60)
 [<b1004b3b>] do_trap+0x8a/0xdb (32)
 [<b1005589>] do_invalid_op+0xae/0xb8 (192)
 [<b1003f97>] error_code+0x4f/0x54 (148)
 [<b103ce18>] posix_cpu_nsleep+0xfd/0x23b (220)
 [<b103969f>] sys_clock_nanosleep+0xe7/0xee (44)
 [<b10033e4>] syscall_call+0x7/0xb (-4020)
Code: 68 00 00 00 00 c7 45 d8 fd ff ff ff e9 db fd ff ff c7 41 6c ff ff ff ff c7 45 d8 01 00 00 00 c7 45 e4 01 00 00 00 e9 e8 fb ff ff <0f> 0b e9 8f fb ff ff b8 80 7a 5a b1 e8 ff cc 3b 00 e9 81 fd ff 
EIP: [<b103cbaa>] posix_cpu_timer_set+0x505/0x52e SS:ESP 0068:c9380e54

<peek>
(gdb) list *posix_cpu_timer_set+0x505
0xb103cbaa is in posix_cpu_timer_set (posix-cpu-timers.c:724).
719 }
720
721 /*
722 * Disarm any old timer after extracting its expiry time.
723 */
724 BUG_ON(!irqs_disabled());
725

726             ret = 0;
727             spin_lock(&p->sighand->siglock);
728             old_expires = timer->it.cpu.expires;

(gdb) list *posix_cpu_nsleep+0xfd
0xb103ce18 is in posix_cpu_nsleep (posix-cpu-timers.c:1597).
1592 static struct itimerspec zero_it;
1593 struct itimerspec it = { .it_value = *rqtp,
1594 .it_interval = {} };
1595

Quoted 8 lines
1596                    spin_lock_irq(&timer.it_lock);
1597                    error = posix_cpu_timer_set(&timer, flags, &it, NULL);
1598                    if (error) {

1599 spin_unlock_irq(&timer.it_lock);
1600 return error;
1601 }
(gdb)
<nope, definitely not Kansas>

7796  ...:0 61481.339ms: user_trace_stop+0xe/0x39c  <= (posix_cpu_timer_set+0x31f/0x547) <--at bug_on() time
7796  ...:0 61481.338ms: rt_lock+0x8/0x29  <= (rt_read_lock+0x33/0x46)
7796  D..:1 61481.338ms: _raw_spin_unlock_irqrestore+0xc/0x50  <= (rt_read_lock+0x2c/0x46)
7796  ...:0 61481.337ms: _raw_spin_lock_irqsave+0xd/0x63  <= (rt_read_lock+0x10/0x46)
7796  ...:0 61481.337ms: rt_read_lock+0x9/0x46  <= (posix_cpu_timer_set+0x7f/0x547)
7796  ...:0 61481.337ms: posix_cpu_timer_set+0xe/0x547  <= (posix_cpu_nsleep+0x106/0x251)
7796  ...:0 61481.336ms: rt_lock+0x8/0x29  <= (posix_cpu_nsleep+0xe2/0x251)
7796  ...:1 61481.336ms: rt_mutex_unlock+0xd/0x37  <= (rt_up+0x32/0x59)
7796  ...:0 61481.335ms: rt_up+0xc/0x59  <= (user_trace_start+0x104/0x1dc)
7796  D..:0 61481.335ms: user_trace_start+0xdd/0x1dc  <= (posix_cpu_nsleep+0x161/0x251) <--once, at first use

OK, it's dying on the very first call, with absolutely nothing between spin_lock_irq() and BUG_ON(!irqs_disabled()), but the spin_lock_irq() has become rt_lock(). Is the BUG_ON() check bogus for the rt kernel?

-Mike

#7 lore raw
Quoted 3 lines · Mike Galbraith

OK, it's dying on the very first call, with absolutely nothing between spin_lock_irq() and BUG_ON(!irqs_disabled()), but the spin_lock_irq() has become rt_lock(). Is the BUG_ON() check bogus for the rt kernel?

Yes. The patch below should help.

tglx

kernel/posix-cpu-timers.c +2 -2
Index: linux-2.6.17-rc6/kernel/posix-cpu-timers.c
===================================================================
@@ -564,7 +564,7 @@
 		p->cpu_timers : p->signal->cpu_timers);
 	head += CPUCLOCK_WHICH(timer->it_clock);
 
-	BUG_ON(!irqs_disabled());
+	BUG_ON_NONRT(!irqs_disabled());
 	spin_lock(&p->sighand->siglock);
 
 	listpos = head;
@@ -721,7 +721,7 @@
 	/*
 	 * Disarm any old timer after extracting its expiry time.
 	 */
-	BUG_ON(!irqs_disabled());
+	BUG_ON_NONRT(!irqs_disabled());
 
 	ret = 0;
 	spin_lock(&p->sighand->siglock);
#8 lore raw

On 6/11/06, Thomas Gleixner <[email protected]> wrote:

Just a quick note to say that 2.6.17-rc6-rt3 booted fine on my AMD64 box. Unlike rc6-rt1 this one will run X with no problems so far. Audio is up, 1394 hard drives are fine, no xruns using Jack at 64/2. I'll report back more if I run into any issues.

Great work!

mark@lightning ~ $ uname -a
Linux lightning 2.6.17-rc6-rt3 #4 PREEMPT Sun Jun 11 12:18:09 PDT 2006 x86_64 AMD Athlon(tm) 64 Processor 3000+ GNU/Linux mark@lightning ~ $

Cheers,
Mark

#9 lore raw
In reply to Ingo Molnar
Quoted 13 lines · Ingo Molnar

i have released the 2.6.17-rc6-rt3 tree, which can be downloaded from the usual place:

http://redhat.com/~mingo/realtime-preempt/

this is a fixes-only release: lots of fixes from Thomas Gleixner (for the softirq problem that caused those ping latency weirdnesses, for hrtimers and timers problems and for the RCU related bug that was causing instability and more), John Stultz, Jan Altenberg and Clark Williams. MIPS update from Manish Lachwani. Futex fix from Dinakar Guniguntala. It also includes the RT-scheduling SMP fix that could fix the scheduling problem reported by Darren Hart.

Quoted 3 lines

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

Great, boots fine on my dual Xeon and solves the ping problem I was
having.

Thomas, any hint at what was going on?

Thanks,

Sébastien.

#10 lore raw
Quoted 9 lines · Sébastien Dugué
Quoted 4 lines

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

Great, boots fine on my dual Xeon and solves the ping problem I was
having.

Thomas, any hint at what was going on?

I missed some modificatons in the networking code when I did the forward to 2.6.17-rc6. The network softirq was raised, but the thread not woken up.

tglx

#11 lore raw
In reply to Sébastien Dugué
Quoted 7 lines · Sébastien Dugué
Quoted 2 lines

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

Great, boots fine on my dual Xeon and solves the ping problem I was
having.

Thomas, any hint at what was going on?

the problem was caused by a mismerge of the __raise_softirq_irqoff() changes of preempt-softirqs. In PREEMPT_SOFTIRQS, softirq activation means a wakeup of the softirq thread - hence __raise_softirq_irqoff() must wake up the softirq thead too. This didnt happen in -rt1 so the network softirq (which processes things like ping reply packets) got delayed to the natural softirq event - the next timer interrupt in the usual case. Hence depending on HZ you got a delay of 1-4-10 msecs (divided into two parts).

Ingo

#12 lore raw
Quoted 10 lines · Ingo Molnar
Quoted 7 lines · Sébastien Dugué
Quoted 2 lines

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

Great, boots fine on my dual Xeon and solves the ping problem I was
having.

Thomas, any hint at what was going on?

Quoted 9 lines

the problem was caused by a mismerge of the __raise_softirq_irqoff() changes of preempt-softirqs. In PREEMPT_SOFTIRQS, softirq activation means a wakeup of the softirq thread - hence __raise_softirq_irqoff() must wake up the softirq thead too. This didnt happen in -rt1 so the network softirq (which processes things like ping reply packets) got delayed to the natural softirq event - the next timer interrupt in the usual case. Hence depending on HZ you got a delay of 1-4-10 msecs (divided into two parts).

Thanks.

Sébastien.

#13 lore raw
In reply to Ingo Molnar
Quoted 16 lines · Ingo Molnar

i have released the 2.6.17-rc6-rt3 tree, which can be downloaded from the usual place:

http://redhat.com/~mingo/realtime-preempt/

this is a fixes-only release: lots of fixes from Thomas Gleixner (for the softirq problem that caused those ping latency weirdnesses, for hrtimers and timers problems and for the RCU related bug that was causing instability and more), John Stultz, Jan Altenberg and Clark Williams. MIPS update from Manish Lachwani. Futex fix from Dinakar Guniguntala. It also includes the RT-scheduling SMP fix that could fix the scheduling problem reported by Darren Hart.

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

Quoted 7 lines

to build a 2.6.17-rc6-rt3 tree, the following patches should be applied:

http://kernel.org/pub/linux/kernel/v2.6/linux-2.6.16.tar.bz2
http://kernel.org/pub/linux/kernel/v2.6/testing/patch-2.6.17-rc6.bz2
http://redhat.com/~mingo/realtime-preempt/patch-2.6.17-rc6-rt3

Ingo

This one still doesn't boot for me on a dual Xeon 2.60. Config is attached and oops is included below.

*****************************************************************************
*

    *
*  REMINDER, the following debugging option is turned on in your
.config:   *
*
    *
*        CONFIG_DEBUG_RT_MUTEXES
     *
*
    *
*  it may increase runtime overhead and latencies.
    *
*
    *

*****************************************************************************
Freeing unused kernel memory: 200k freed
input: ImExPS/2 Generic Explorer Mouse as /class/input/input1 kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.

BUG: unable to handle kernel paging request at virtual address f3010000
 printing eip:
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    1
EIP:    0060:[<c0132f9c>]    Not tainted VLI
EFLAGS: 00010297   (2.6.17-rc6-rt4 #10)

EIP is at lookup_symbol+0x11/0x35
eax: 00000001 ebx: e083185c ecx: c02f20c4 edx: c02f0000
esi: f3010000 edi: e083185c ebp: df597e80 esp: df597e74
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process modprobe (pid: 1419, threadinfo=df596000 task=dec3ac90 stack_left=7744 worst_left=-1)

Stack: e083b580 00000bf0 e083185c df597e9c c0132fe5 df597eb4 df597eb0
e083b580
       00000bf0 e083185c df597ec4 c0133c93 00000001 00000012 e082dde8
00000000
       df597ed8 e0839200 00000bf0 e082dde8 df597ee8 c01341fa e083b580
00000000
Call Trace:
 [<c01036a1>] show_stack_log_lvl+0x82/0x8a (36)
 [<c0103821>] show_registers+0x139/0x1a1 (32)
 [<c0103a15>] die+0x118/0x1df (60)
 [<c0110cf3>] do_page_fault+0x45c/0x532 (76)
 [<c010336b>] error_code+0x4f/0x54 (72)
 [<c0132fe5>] __find_symbol+0x25/0x1b7 (28)
 [<c0133c93>] resolve_symbol+0x27/0x5f (40)
 [<c01341fa>] simplify_symbols+0x83/0xf3 (36)
 [<c0134e31>] load_module+0x668/0x9e2 (184)
 [<c0135210>] sys_init_module+0x42/0x1a4 (20)
 [<c01027fb>] sysenter_past_esp+0x54/0x75 (-8116)
Code: eb 11 8b 75 f0 41 83 c2 28 0f b7 46 30 39 c1 72 c9 31 c0 5a 59 5b

5e 5f 5d c3 55 89 e5 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 04 89
EIP: [<c0132f9c>] lookup_symbol+0x11/0x35 SS:ESP 0068:df597e74

Signature · 1 line
   kr
#14 lore raw
Quoted 82 lines · K.R. Foley
Quoted 16 lines · Ingo Molnar

i have released the 2.6.17-rc6-rt3 tree, which can be downloaded from the usual place:

http://redhat.com/~mingo/realtime-preempt/

this is a fixes-only release: lots of fixes from Thomas Gleixner (for the softirq problem that caused those ping latency weirdnesses, for hrtimers and timers problems and for the RCU related bug that was causing instability and more), John Stultz, Jan Altenberg and Clark Williams. MIPS update from Manish Lachwani. Futex fix from Dinakar Guniguntala. It also includes the RT-scheduling SMP fix that could fix the scheduling problem reported by Darren Hart.

I think all of the regressions reported against rt1 are fixed, please re-report if any of them is still unfixed.

Quoted 7 lines

to build a 2.6.17-rc6-rt3 tree, the following patches should be applied:

http://kernel.org/pub/linux/kernel/v2.6/linux-2.6.16.tar.bz2
http://kernel.org/pub/linux/kernel/v2.6/testing/patch-2.6.17-rc6.bz2
http://redhat.com/~mingo/realtime-preempt/patch-2.6.17-rc6-rt3

Ingo

This one still doesn't boot for me on a dual Xeon 2.60. Config is attached and oops is included below.

*****************************************************************************
*

    *
*  REMINDER, the following debugging option is turned on in your
.config:   *
*
    *
*        CONFIG_DEBUG_RT_MUTEXES
     *
*
    *
*  it may increase runtime overhead and latencies.
    *
*
    *

*****************************************************************************
Freeing unused kernel memory: 200k freed
input: ImExPS/2 Generic Explorer Mouse as /class/input/input1 kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.

BUG: unable to handle kernel paging request at virtual address f3010000
 printing eip:
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    1
EIP:    0060:[<c0132f9c>]    Not tainted VLI
EFLAGS: 00010297   (2.6.17-rc6-rt4 #10)

EIP is at lookup_symbol+0x11/0x35
eax: 00000001 ebx: e083185c ecx: c02f20c4 edx: c02f0000
esi: f3010000 edi: e083185c ebp: df597e80 esp: df597e74
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process modprobe (pid: 1419, threadinfo=df596000 task=dec3ac90 stack_left=7744 worst_left=-1)

Stack: e083b580 00000bf0 e083185c df597e9c c0132fe5 df597eb4 df597eb0
e083b580
       00000bf0 e083185c df597ec4 c0133c93 00000001 00000012 e082dde8
00000000
       df597ed8 e0839200 00000bf0 e082dde8 df597ee8 c01341fa e083b580
00000000
Call Trace:
 [<c01036a1>] show_stack_log_lvl+0x82/0x8a (36)
 [<c0103821>] show_registers+0x139/0x1a1 (32)
 [<c0103a15>] die+0x118/0x1df (60)
 [<c0110cf3>] do_page_fault+0x45c/0x532 (76)
 [<c010336b>] error_code+0x4f/0x54 (72)
 [<c0132fe5>] __find_symbol+0x25/0x1b7 (28)
 [<c0133c93>] resolve_symbol+0x27/0x5f (40)
 [<c01341fa>] simplify_symbols+0x83/0xf3 (36)
 [<c0134e31>] load_module+0x668/0x9e2 (184)
 [<c0135210>] sys_init_module+0x42/0x1a4 (20)
 [<c01027fb>] sysenter_past_esp+0x54/0x75 (-8116)
Code: eb 11 8b 75 f0 41 83 c2 28 0f b7 46 30 39 c1 72 c9 31 c0 5a 59 5b
Quoted 7 lines

5e 5f 5d c3 55 89 e5 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 04 89
EIP: [<c0132f9c>] lookup_symbol+0x11/0x35 SS:ESP 0068:df597e74

DOH! That was actually 2.6.17-rc6-rt4. Sorry.

Signature · 1 line
   kr
#15 lore raw
Quoted 38 lines · K.R. Foley
Quoted 34 lines
BUG: unable to handle kernel paging request at virtual address f3010000
 printing eip:
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    1
EIP:    0060:[<c0132f9c>]    Not tainted VLI
EFLAGS: 00010297   (2.6.17-rc6-rt4 #10)

EIP is at lookup_symbol+0x11/0x35
eax: 00000001 ebx: e083185c ecx: c02f20c4 edx: c02f0000
esi: f3010000 edi: e083185c ebp: df597e80 esp: df597e74
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process modprobe (pid: 1419, threadinfo=df596000 task=dec3ac90 stack_left=7744 worst_left=-1)

Stack: e083b580 00000bf0 e083185c df597e9c c0132fe5 df597eb4 df597eb0
e083b580
       00000bf0 e083185c df597ec4 c0133c93 00000001 00000012 e082dde8
00000000
       df597ed8 e0839200 00000bf0 e082dde8 df597ee8 c01341fa e083b580
00000000
Call Trace:
 [<c01036a1>] show_stack_log_lvl+0x82/0x8a (36)
 [<c0103821>] show_registers+0x139/0x1a1 (32)
 [<c0103a15>] die+0x118/0x1df (60)
 [<c0110cf3>] do_page_fault+0x45c/0x532 (76)
 [<c010336b>] error_code+0x4f/0x54 (72)
 [<c0132fe5>] __find_symbol+0x25/0x1b7 (28)
 [<c0133c93>] resolve_symbol+0x27/0x5f (40)
 [<c01341fa>] simplify_symbols+0x83/0xf3 (36)
 [<c0134e31>] load_module+0x668/0x9e2 (184)
 [<c0135210>] sys_init_module+0x42/0x1a4 (20)
 [<c01027fb>] sysenter_past_esp+0x54/0x75 (-8116)
Code: eb 11 8b 75 f0 41 83 c2 28 0f b7 46 30 39 c1 72 c9 31 c0 5a 59 5b
Quoted 3 lines

5e 5f 5d c3 55 89 e5 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 04 89
EIP: [<c0132f9c>] lookup_symbol+0x11/0x35 SS:ESP 0068:df597e74

Quoted 1 line

DOH! That was actually 2.6.17-rc6-rt4. Sorry.

Which module is it trying to load ?

tglx

#16 lore raw
Quoted 38 lines · Thomas Gleixner
Quoted 37 lines · K.R. Foley
Quoted 37 lines
BUG: unable to handle kernel paging request at virtual address f3010000
 printing eip:
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    1
EIP:    0060:[<c0132f9c>]    Not tainted VLI
EFLAGS: 00010297   (2.6.17-rc6-rt4 #10)

EIP is at lookup_symbol+0x11/0x35
eax: 00000001 ebx: e083185c ecx: c02f20c4 edx: c02f0000
esi: f3010000 edi: e083185c ebp: df597e80 esp: df597e74
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process modprobe (pid: 1419, threadinfo=df596000 task=dec3ac90 stack_left=7744 worst_left=-1)

Stack: e083b580 00000bf0 e083185c df597e9c c0132fe5 df597eb4 df597eb0
e083b580
       00000bf0 e083185c df597ec4 c0133c93 00000001 00000012 e082dde8
00000000
       df597ed8 e0839200 00000bf0 e082dde8 df597ee8 c01341fa e083b580
00000000
Call Trace:
 [<c01036a1>] show_stack_log_lvl+0x82/0x8a (36)
 [<c0103821>] show_registers+0x139/0x1a1 (32)
 [<c0103a15>] die+0x118/0x1df (60)
 [<c0110cf3>] do_page_fault+0x45c/0x532 (76)
 [<c010336b>] error_code+0x4f/0x54 (72)
 [<c0132fe5>] __find_symbol+0x25/0x1b7 (28)
 [<c0133c93>] resolve_symbol+0x27/0x5f (40)
 [<c01341fa>] simplify_symbols+0x83/0xf3 (36)
 [<c0134e31>] load_module+0x668/0x9e2 (184)
 [<c0135210>] sys_init_module+0x42/0x1a4 (20)
 [<c01027fb>] sysenter_past_esp+0x54/0x75 (-8116)
Code: eb 11 8b 75 f0 41 83 c2 28 0f b7 46 30 39 c1 72 c9 31 c0 5a 59 5b

5e 5f 5d c3 55 89 e5 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 04 89
EIP: [<c0132f9c>] lookup_symbol+0x11/0x35 SS:ESP 0068:df597e74

Quoted 5 lines
Quoted 1 line

DOH! That was actually 2.6.17-rc6-rt4. Sorry.

Which module is it trying to load ?

tglx

Can't really say which it is trying to load when it dies. The lines below are the lines that immediately preceed the oops.

NET: Registered protocol family 1
input: AT Translated Set 2 keyboard as /class/input/input0
NET: Registered protocol family 17
NET: Registered protocol family 8
NET: Registered protocol family 20
Starting balanced_irq
Using IPI Shortcut mode
Time: tsc clocksource has been installed.
hrtimers: Switched to high resolution mode CPU 3
hrtimers: Switched to high resolution mode CPU 2
hrtimers: Switched to high resolution mode CPU 1
hrtimers: Switched to high resolution mode CPU 0
*****************************************************************************
*

    *
*  REMINDER, the following debugging option is turned on in your
.config:   *
*
    *
*        CONFIG_DEBUG_RT_MUTEXES
     *
*
    *
*  it may increase runtime overhead and latencies.
    *
*
    *

*****************************************************************************
Freeing unused kernel memory: 200k freed
input: ImExPS/2 Generic Explorer Mouse as /class/input/input1 kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.

Signature · 1 line
   kr
#17 lore raw
Quoted 79 lines · K.R. Foley
Quoted 40 lines · Thomas Gleixner
Quoted 38 lines · K.R. Foley
Quoted 38 lines
BUG: unable to handle kernel paging request at virtual address f3010000
 printing eip:
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    1
EIP:    0060:[<c0132f9c>]    Not tainted VLI
EFLAGS: 00010297   (2.6.17-rc6-rt4 #10)

EIP is at lookup_symbol+0x11/0x35
eax: 00000001 ebx: e083185c ecx: c02f20c4 edx: c02f0000
esi: f3010000 edi: e083185c ebp: df597e80 esp: df597e74
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process modprobe (pid: 1419, threadinfo=df596000 task=dec3ac90 stack_left=7744 worst_left=-1)

Stack: e083b580 00000bf0 e083185c df597e9c c0132fe5 df597eb4 df597eb0
e083b580
       00000bf0 e083185c df597ec4 c0133c93 00000001 00000012 e082dde8
00000000
       df597ed8 e0839200 00000bf0 e082dde8 df597ee8 c01341fa e083b580
00000000
Call Trace:
 [<c01036a1>] show_stack_log_lvl+0x82/0x8a (36)
 [<c0103821>] show_registers+0x139/0x1a1 (32)
 [<c0103a15>] die+0x118/0x1df (60)
 [<c0110cf3>] do_page_fault+0x45c/0x532 (76)
 [<c010336b>] error_code+0x4f/0x54 (72)
 [<c0132fe5>] __find_symbol+0x25/0x1b7 (28)
 [<c0133c93>] resolve_symbol+0x27/0x5f (40)
 [<c01341fa>] simplify_symbols+0x83/0xf3 (36)
 [<c0134e31>] load_module+0x668/0x9e2 (184)
 [<c0135210>] sys_init_module+0x42/0x1a4 (20)
 [<c01027fb>] sysenter_past_esp+0x54/0x75 (-8116)
Code: eb 11 8b 75 f0 41 83 c2 28 0f b7 46 30 39 c1 72 c9 31 c0 5a 59 5b

5e 5f 5d c3 55 89 e5 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 04 89
EIP: [<c0132f9c>] lookup_symbol+0x11/0x35 SS:ESP 0068:df597e74

Quoted 6 lines
Quoted 2 lines

DOH! That was actually 2.6.17-rc6-rt4. Sorry.

Which module is it trying to load ?

tglx

Can't really say which it is trying to load when it dies. The lines below are the lines that immediately preceed the oops.

NET: Registered protocol family 1
input: AT Translated Set 2 keyboard as /class/input/input0
NET: Registered protocol family 17
NET: Registered protocol family 8
NET: Registered protocol family 20
Starting balanced_irq
Using IPI Shortcut mode
Time: tsc clocksource has been installed.
hrtimers: Switched to high resolution mode CPU 3
hrtimers: Switched to high resolution mode CPU 2
hrtimers: Switched to high resolution mode CPU 1
hrtimers: Switched to high resolution mode CPU 0
*****************************************************************************
*

    *
*  REMINDER, the following debugging option is turned on in your
.config:   *
*
    *
*        CONFIG_DEBUG_RT_MUTEXES
     *
*
    *
*  it may increase runtime overhead and latencies.
    *
*
    *
Quoted 7 lines

*****************************************************************************
Freeing unused kernel memory: 200k freed
input: ImExPS/2 Generic Explorer Mouse as /class/input/input1 kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.

Similar for me on a dual Xeon 3.2 with 2.6.17-rc6-rt4:

*****************************************************************************
Time: tsc clocksource has been installed.
hrtimers: Switched to high resolution mode CPU 0
hrtimers: Switched to high resolution mode CPU 1

*                                                                           *
*  REMINDER, the following debugging option is turned on in your .config:   *
*                                                                           *
*        CONFIG_DEBUG_RT_MUTEXES                                             *
*                                                                           *
*  it may increase runtime overhead and latencies.                          *
*                                                                           *

*****************************************************************************
Freeing unused kernel memory: 208k freed
Red Hat nash version 4.1.18.1 starting
Mounted /proc filesystem
Mounting sysfs
Creating /dev
Starting udev
Loading jbd.ko mBUG: unable to handle kernel paging request at virtual address 75010000 printing eip:
c0135679
*pde = 00000000

Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:

CPU: 1
Eodule
IP: 0060:[<c0135679>] Not tainted VLI
EFLAGS: 00010297 (2.6.17-rc6-rt5 #1)
EIP is at lookup_symbol+0xe/0x31
eax: ffffffff ebx: f881a7d2 ecx: c0332f3c edx: c03309d8
esi: 75010000 edi: f881a7d2 ebp: f7f27ec0 esp: f7f27e8c
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process insmod (pid: 293, threadinfo=f7f27000 task=f7f360f0 stack_left=3672 worst_left=-1)

Stack: f882e520 000010a0 f881a7d2 c01356bd f7f27ebc f882e520 000010a0 f881a7d2
      00000012 c013630c 00000001 f881752c 00000000 c0334284 f882abc0 
000010a0
      f881752c 0000008f c0136859 f882e520 00000000 f881931c f882e52c 
f882e52d
Call Trace:
[<c01356bd>] __find_symbol+0x21/0x1b3 (16)
[<c013630c>] resolve_symbol+0x27/0x61 (24)
[<c0136859>] simplify_symbols+0x85/0xf7 (36)
[<c0137532>] load_module+0x73f/0xaf9 (32)
[<c013373e>] try_to_take_rt_mutex+0x165/0x172 (20)
[<c013792f>] sys_init_module+0x24/0x1a0 (16)
[<c013794d>] sys_init_module+0x42/0x1a0 (144)
[<c01032d3>] sysenter_past_esp+0x54/0x75 (16)
Code: 01 85 c0 75 04 89 c8 eb 0e 0f b7 45 30 41 83 c2 28 39 c1 72 cb 31 c0 5a 5b 5e 5f 5d c3 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0
#18 lore raw
Quoted 127 lines · Bruce Eleniak
Quoted 80 lines · K.R. Foley
Quoted 40 lines · Thomas Gleixner
Quoted 38 lines · K.R. Foley
Quoted 38 lines
BUG: unable to handle kernel paging request at virtual address f3010000
 printing eip:
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    1
EIP:    0060:[<c0132f9c>]    Not tainted VLI
EFLAGS: 00010297   (2.6.17-rc6-rt4 #10)

EIP is at lookup_symbol+0x11/0x35
eax: 00000001 ebx: e083185c ecx: c02f20c4 edx: c02f0000
esi: f3010000 edi: e083185c ebp: df597e80 esp: df597e74
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process modprobe (pid: 1419, threadinfo=df596000 task=dec3ac90 stack_left=7744 worst_left=-1)

Stack: e083b580 00000bf0 e083185c df597e9c c0132fe5 df597eb4 df597eb0
e083b580
       00000bf0 e083185c df597ec4 c0133c93 00000001 00000012 e082dde8
00000000
       df597ed8 e0839200 00000bf0 e082dde8 df597ee8 c01341fa e083b580
00000000
Call Trace:
 [<c01036a1>] show_stack_log_lvl+0x82/0x8a (36)
 [<c0103821>] show_registers+0x139/0x1a1 (32)
 [<c0103a15>] die+0x118/0x1df (60)
 [<c0110cf3>] do_page_fault+0x45c/0x532 (76)
 [<c010336b>] error_code+0x4f/0x54 (72)
 [<c0132fe5>] __find_symbol+0x25/0x1b7 (28)
 [<c0133c93>] resolve_symbol+0x27/0x5f (40)
 [<c01341fa>] simplify_symbols+0x83/0xf3 (36)
 [<c0134e31>] load_module+0x668/0x9e2 (184)
 [<c0135210>] sys_init_module+0x42/0x1a4 (20)
 [<c01027fb>] sysenter_past_esp+0x54/0x75 (-8116)
Code: eb 11 8b 75 f0 41 83 c2 28 0f b7 46 30 39 c1 72 c9 31 c0 5a 59 5b

5e 5f 5d c3 55 89 e5 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 04 89
EIP: [<c0132f9c>] lookup_symbol+0x11/0x35 SS:ESP 0068:df597e74

Quoted 6 lines
Quoted 2 lines

DOH! That was actually 2.6.17-rc6-rt4. Sorry.

Which module is it trying to load ?

tglx

Can't really say which it is trying to load when it dies. The lines below are the lines that immediately preceed the oops.

NET: Registered protocol family 1
input: AT Translated Set 2 keyboard as /class/input/input0
NET: Registered protocol family 17
NET: Registered protocol family 8
NET: Registered protocol family 20
Starting balanced_irq
Using IPI Shortcut mode
Time: tsc clocksource has been installed.
hrtimers: Switched to high resolution mode CPU 3
hrtimers: Switched to high resolution mode CPU 2
hrtimers: Switched to high resolution mode CPU 1
hrtimers: Switched to high resolution mode CPU 0
*****************************************************************************

*

    *
*  REMINDER, the following debugging option is turned on in your
.config:   *
*
    *
*        CONFIG_DEBUG_RT_MUTEXES
     *
*
    *
*  it may increase runtime overhead and latencies.
    *
*
    *
Quoted 8 lines

*****************************************************************************

Freeing unused kernel memory: 200k freed
input: ImExPS/2 Generic Explorer Mouse as /class/input/input1 kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.

Similar for me on a dual Xeon 3.2 with 2.6.17-rc6-rt4:

*****************************************************************************

Time: tsc clocksource has been installed.
hrtimers: Switched to high resolution mode CPU 0
hrtimers: Switched to high resolution mode CPU 1

*                                                                           *
*  REMINDER, the following debugging option is turned on in your .config:   *
*                                                                           *
*        CONFIG_DEBUG_RT_MUTEXES                                             *
*                                                                           *
*  it may increase runtime overhead and latencies.                          *
*                                                                           *

*****************************************************************************

Freeing unused kernel memory: 208k freed
Red Hat nash version 4.1.18.1 starting
Mounted /proc filesystem
Mounting sysfs
Creating /dev
Starting udev
Loading jbd.ko mBUG: unable to handle kernel paging request at virtual address 75010000 printing eip:
c0135679
*pde = 00000000

Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:

CPU: 1
Eodule
IP: 0060:[<c0135679>] Not tainted VLI
EFLAGS: 00010297 (2.6.17-rc6-rt5 #1)
EIP is at lookup_symbol+0xe/0x31
eax: ffffffff ebx: f881a7d2 ecx: c0332f3c edx: c03309d8
esi: 75010000 edi: f881a7d2 ebp: f7f27ec0 esp: f7f27e8c
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process insmod (pid: 293, threadinfo=f7f27000 task=f7f360f0 stack_left=3672 worst_left=-1)

Quoted 17 lines
Stack: f882e520 000010a0 f881a7d2 c01356bd f7f27ebc f882e520 000010a0 f881a7d2
      00000012 c013630c 00000001 f881752c 00000000 c0334284 f882abc0 
000010a0
      f881752c 0000008f c0136859 f882e520 00000000 f881931c f882e52c 
f882e52d
Call Trace:
[<c01356bd>] __find_symbol+0x21/0x1b3 (16)
[<c013630c>] resolve_symbol+0x27/0x61 (24)
[<c0136859>] simplify_symbols+0x85/0xf7 (36)
[<c0137532>] load_module+0x73f/0xaf9 (32)
[<c013373e>] try_to_take_rt_mutex+0x165/0x172 (20)
[<c013792f>] sys_init_module+0x24/0x1a0 (16)
[<c013794d>] sys_init_module+0x42/0x1a0 (144)
[<c01032d3>] sysenter_past_esp+0x54/0x75 (16)
Code: 01 85 c0 75 04 89 c8 eb 0e 0f b7 45 30 41 83 c2 28 39 c1 72 cb 31 c0 5a 5b 5e 5f 5d c3 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0

Whoops. 2.6.17-rc6-rt5 sorry. FWIW, Thomas' 2.6.17-rc6-hrt4 runs fine in isolation from the rt patchset.

#19 lore raw
Quoted 160 lines · Bruce Eleniak
Quoted 155 lines · Bruce Eleniak
Quoted 83 lines · K.R. Foley
Quoted 43 lines · Thomas Gleixner
Quoted 41 lines · K.R. Foley
Quoted 41 lines
BUG: unable to handle kernel paging request at virtual address
f3010000
 printing eip:
*pde = 00000000
Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:
CPU:    1
EIP:    0060:[<c0132f9c>]    Not tainted VLI
EFLAGS: 00010297   (2.6.17-rc6-rt4 #10)

EIP is at lookup_symbol+0x11/0x35
eax: 00000001 ebx: e083185c ecx: c02f20c4 edx: c02f0000
esi: f3010000 edi: e083185c ebp: df597e80 esp: df597e74
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process modprobe (pid: 1419, threadinfo=df596000 task=dec3ac90 stack_left=7744 worst_left=-1)

Stack: e083b580 00000bf0 e083185c df597e9c c0132fe5 df597eb4 df597eb0
e083b580
       00000bf0 e083185c df597ec4 c0133c93 00000001 00000012 e082dde8
00000000
       df597ed8 e0839200 00000bf0 e082dde8 df597ee8 c01341fa e083b580
00000000
Call Trace:
 [<c01036a1>] show_stack_log_lvl+0x82/0x8a (36)
 [<c0103821>] show_registers+0x139/0x1a1 (32)
 [<c0103a15>] die+0x118/0x1df (60)
 [<c0110cf3>] do_page_fault+0x45c/0x532 (76)
 [<c010336b>] error_code+0x4f/0x54 (72)
 [<c0132fe5>] __find_symbol+0x25/0x1b7 (28)
 [<c0133c93>] resolve_symbol+0x27/0x5f (40)
 [<c01341fa>] simplify_symbols+0x83/0xf3 (36)
 [<c0134e31>] load_module+0x668/0x9e2 (184)
 [<c0135210>] sys_init_module+0x42/0x1a4 (20)
 [<c01027fb>] sysenter_past_esp+0x54/0x75 (-8116)
Code: eb 11 8b 75 f0 41 83 c2 28 0f b7 46 30 39 c1 72 c9 31 c0 5a

59 5b
5e 5f 5d c3 55 89 e5 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae
75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0 75 04 89
EIP: [<c0132f9c>] lookup_symbol+0x11/0x35 SS:ESP 0068:df597e74

Quoted 6 lines
Quoted 2 lines

DOH! That was actually 2.6.17-rc6-rt4. Sorry.

Which module is it trying to load ?

tglx

Can't really say which it is trying to load when it dies. The lines below are the lines that immediately preceed the oops.

NET: Registered protocol family 1
input: AT Translated Set 2 keyboard as /class/input/input0
NET: Registered protocol family 17
NET: Registered protocol family 8
NET: Registered protocol family 20
Starting balanced_irq
Using IPI Shortcut mode
Time: tsc clocksource has been installed.
hrtimers: Switched to high resolution mode CPU 3
hrtimers: Switched to high resolution mode CPU 2
hrtimers: Switched to high resolution mode CPU 1
hrtimers: Switched to high resolution mode CPU 0
*****************************************************************************

*

    *
*  REMINDER, the following debugging option is turned on in your
.config:   *
*
    *
*        CONFIG_DEBUG_RT_MUTEXES
     *
*
    *
*  it may increase runtime overhead and latencies.
    *
*
    *
Quoted 8 lines

*****************************************************************************

Freeing unused kernel memory: 200k freed
input: ImExPS/2 Generic Explorer Mouse as /class/input/input1 kjournald starting. Commit interval 5 seconds
EXT3-fs: mounted filesystem with ordered data mode.

Similar for me on a dual Xeon 3.2 with 2.6.17-rc6-rt4:

*****************************************************************************

Time: tsc clocksource has been installed.
hrtimers: Switched to high resolution mode CPU 0
hrtimers: Switched to high resolution mode CPU 1
*
*
* REMINDER, the following debugging option is turned on in your
.config: *
*
*
*
CONFIG_DEBUG_RT_MUTEXES *
*
*
* it may increase runtime overhead and
latencies. *
*
*
*****************************************************************************

Freeing unused kernel memory: 208k freed
Red Hat nash version 4.1.18.1 starting
Mounted /proc filesystem
Mounting sysfs
Creating /dev
Starting udev
Loading jbd.ko mBUG: unable to handle kernel paging request at virtual address 75010000
printing eip:
c0135679
*pde = 00000000

Oops: 0000 [#1]
PREEMPT SMP
Modules linked in:

CPU: 1
Eodule
IP: 0060:[<c0135679>] Not tainted VLI
EFLAGS: 00010297 (2.6.17-rc6-rt5 #1)
EIP is at lookup_symbol+0xe/0x31
eax: ffffffff ebx: f881a7d2 ecx: c0332f3c edx: c03309d8
esi: 75010000 edi: f881a7d2 ebp: f7f27ec0 esp: f7f27e8c
ds: 007b es: 007b ss: 0068 preempt: 00000001
Process insmod (pid: 293, threadinfo=f7f27000 task=f7f360f0 stack_left=3672 worst_left=-1)

Stack: f882e520 000010a0 f881a7d2 c01356bd f7f27ebc f882e520 000010a0
f881a7d2
      00000012 c013630c 00000001 f881752c 00000000 c0334284 f882abc0
000010a0
      f881752c 0000008f c0136859 f882e520 00000000 f881931c f882e52c
f882e52d
Call Trace:
[<c01356bd>] __find_symbol+0x21/0x1b3 (16)
[<c013630c>] resolve_symbol+0x27/0x61 (24)
[<c0136859>] simplify_symbols+0x85/0xf7 (36)
[<c0137532>] load_module+0x73f/0xaf9 (32)
[<c013373e>] try_to_take_rt_mutex+0x165/0x172 (20)
[<c013792f>] sys_init_module+0x24/0x1a0 (16)
[<c013794d>] sys_init_module+0x42/0x1a0 (144)
[<c01032d3>] sysenter_past_esp+0x54/0x75 (16)
Code: 01 85 c0 75 04 89 c8 eb 0e 0f b7 45 30 41 83 c2 28 39 c1 72 cb
Quoted 4 lines

31 c0 5a 5b 5e 5f 5d c3 57 56 53 89 c3 39 ca 73 22 8b 72 04 89 df <ac> ae 75 08 84 c0 75 f8 31 c0 eb 04 19 c0 0c 01 85 c0

Quoted 3 lines

Whoops. 2.6.17-rc6-rt5 sorry. FWIW, Thomas' 2.6.17-rc6-hrt4 runs fine in isolation from the rt patchset.

Everything from 2.6.17-rc6-rt3 forward dies for sure on all my SMP boxes with the same error above. I also had problems with 2.6.17-rc6-rt1, but I can't say for sure if they were the same problems. I just haven't had time this week to dig into it much more.

Signature · 1 line
   kr