View previous topic :: View next topic |
Author |
Message |
flameman n00b
Joined: 10 Apr 2011 Posts: 39
|
Posted: Mon Apr 25, 2011 2:34 pm Post subject: on ppc 2xG4 aka "mirrored doors" rcu detected cpu0 |
|
|
hello
i am trying to install 2008-minimal-install-cd-gentoo on "mirrored doors" machine
but I am stopped at the boot stage:
when I hear the BELL sound i pressed "C"
so the CD boots his first stage in where I am asked about which kernel and what kernel-parameters
i simply typed "ppc32"
and i see something booting on my screen
but the kernel does not complete his boot, it stops without an apparently panic, just sayng something about the cpu_core0 ???
I don't get any other messages, so not sure what the problem
is.
Last edited by flameman on Tue Apr 26, 2011 9:12 am; edited 2 times in total |
|
Back to top |
|
|
flameman n00b
Joined: 10 Apr 2011 Posts: 39
|
Posted: Mon Apr 25, 2011 6:16 pm Post subject: |
|
|
ok
i downloaded the last 2010 minimal cd.iso and booted it with "ppc32"
it stops with the following message
info rcu detected cpu0 stall (t=10000 jiffies)
info rcu detected cpu0 stall (t=40000 jiffies)
info rcu detected cpu0 stall (t=80000 jiffies)
....
what does it mean ? |
|
Back to top |
|
|
blob999 n00b
Joined: 07 Jun 2009 Posts: 64 Location: NATO area
|
Posted: Mon Apr 25, 2011 7:25 pm Post subject: |
|
|
it means that the bootkernel has a bug to handle SMP or that your machine has an hw failure.
Try passing, at yaboot prompt:
ppc32 video=ofonly udbg-immortal
and report any results (e.g. with a screen pics). _________________ LinuxPPC user!
my blog: http://linuxpowerpc.blogspot.com/ |
|
Back to top |
|
|
flameman n00b
Joined: 10 Apr 2011 Posts: 39
|
Posted: Mon Apr 25, 2011 11:31 pm Post subject: |
|
|
=(
for what Documentation/RCU/stallwarn.txt has to say about it.
Quote: |
Using RCU's CPU Stall Detector
The CONFIG_RCU_CPU_STALL_DETECTOR kernel config parameter enables
RCU's CPU stall detector, which detects conditions that unduly delay
RCU grace periods. The stall detector's idea of what constitutes
"unduly delayed" is controlled by a set of C preprocessor macros:
RCU_SECONDS_TILL_STALL_CHECK
This macro defines the period of time that RCU will wait from
the beginning of a grace period until it issues an RCU CPU
stall warning. This time period is normally ten seconds.
RCU_SECONDS_TILL_STALL_RECHECK
This macro defines the period of time that RCU will wait after
issuing a stall warning until it issues another stall warning
for the same stall. This time period is normally set to thirty
seconds.
RCU_STALL_RAT_DELAY
The CPU stall detector tries to make the offending CPU print its
own warnings, as this often gives better-quality stack traces.
However, if the offending CPU does not detect its own stall in
the number of jiffies specified by RCU_STALL_RAT_DELAY, then
some other CPU will complain. This delay is normally set to
two jiffies.
When a CPU detects that it is stalling, it will print a message similar
to the following:
INFO: rcu_sched_state detected stall on CPU 5 (t=2500 jiffies)
This message indicates that CPU 5 detected that it was causing a stall,
and that the stall was affecting RCU-sched. This message will normally be
followed by a stack dump of the offending CPU. On TREE_RCU kernel builds,
RCU and RCU-sched are implemented by the same underlying mechanism,
while on TREE_PREEMPT_RCU kernel builds, RCU is instead implemented
by rcu_preempt_state.
On the other hand, if the offending CPU fails to print out a stall-warning
message quickly enough, some other CPU will print a message similar to
the following:
INFO: rcu_bh_state detected stalls on CPUs/tasks: { 3 5 } (detected by 2, 2502 jiffies)
This message indicates that CPU 2 detected that CPUs 3 and 5 were both
causing stalls, and that the stall was affecting RCU-bh. This message
will normally be followed by stack dumps for each CPU. Please note that
TREE_PREEMPT_RCU builds can be stalled by tasks as well as by CPUs,
and that the tasks will be indicated by PID, for example, "P3421".
It is even possible for a rcu_preempt_state stall to be caused by both
CPUs -and- tasks, in which case the offending CPUs and tasks will all
be called out in the list.
Finally, if the grace period ends just as the stall warning starts
printing, there will be a spurious stall-warning message:
INFO: rcu_bh_state detected stalls on CPUs/tasks: { } (detected by 4, 2502 jiffies)
This is rare, but does happen from time to time in real life.
So your kernel printed an RCU CPU stall warning. The next question is
"What caused it?" The following problems can result in RCU CPU stall
warnings:
o A CPU looping in an RCU read-side critical section.
o A CPU looping with interrupts disabled. This condition can
result in RCU-sched and RCU-bh stalls.
o A CPU looping with preemption disabled. This condition can
result in RCU-sched stalls and, if ksoftirqd is in use, RCU-bh
stalls.
o A CPU looping with bottom halves disabled. This condition can
result in RCU-sched and RCU-bh stalls.
o For !CONFIG_PREEMPT kernels, a CPU looping anywhere in the kernel
without invoking schedule().
o A bug in the RCU implementation.
o A hardware failure. This is quite unlikely, but has occurred
at least once in real life. A CPU failed in a running system,
becoming unresponsive, but not causing an immediate crash.
This resulted in a series of RCU CPU stall warnings, eventually
leading the realization that the CPU had failed.
The RCU, RCU-sched, and RCU-bh implementations have CPU stall
warning. SRCU does not have its own CPU stall warnings, but its
calls to synchronize_sched() will result in RCU-sched detecting
RCU-sched-related CPU stalls. Please note that RCU only detects
CPU stalls when there is a grace period in progress. No grace period,
no CPU stall warnings.
To diagnose the cause of the stall, inspect the stack traces.
The offending function will usually be near the top of the stack.
If you have a series of stall warnings from a single extended stall,
comparing the stack traces can often help determine where the stall
is occurring, which will usually be in the function nearest the top of
that portion of the stack which remains the same from trace to trace.
If you can reliably trigger the stall, ftrace can be quite helpful.
RCU bugs can often be debugged with the help of CONFIG_RCU_TRACE.
|
Last edited by flameman on Tue Apr 26, 2011 8:55 am; edited 2 times in total |
|
Back to top |
|
|
flameman n00b
Joined: 10 Apr 2011 Posts: 39
|
Posted: Mon Apr 25, 2011 11:59 pm Post subject: |
|
|
blob999 wrote: |
ppc32 video=ofonly udbg-immortal
|
no progress, the ofonly forces a very slow text framebuffer, while the udbg-immortal does not report or cause any other interesting info: it just makes the screen "dirty" ' disabling the fb refresh, so all the old text lines are not cleared when the screen scroll
... anyway, the kernel is still stopped with
info rcu detected cpu0 stall (t=10000 jiffies)
info rcu detected cpu0 stall (t=40000 jiffies)
info rcu detected cpu0 stall (t=80000 jiffies)
....
may be a kernel bug ?
i tried an old debian liveCD (very old, it has kernel v 2.6.8...), which booted with success, but the cat /proc/cpuinfo is telling me that kernel was compiled without the SMP support (dunno if a so old kernel had SMP, i pretty think no =P), so ... one of the two cpu is pretting working with debian kernel, the problem is around the second cpu, or the kernel smp support, or the hw support for dual cpu
debian-kernel has been compiling for 2 hours without any issue: it does not mean that all the 2 cpu are ok, but it means that at least one of the 2 cpu is ok
how could I investigate both ? I booted morphOS (similar to the amigaOS for PPC, dunno about it, i just found this OS installed where I bought this powermac), which should have SMP support, and he tells there is no problem with cpu0 (ppc7450) and cpu1 (ppc7450)
unfortunately i have no compiler on morphos and no HEAVY application compiled with SMP support, so ... dunno at all, i could only trust the sys-info, which is reporting that the morphos kernel has correctly recognized, initialized, and sheduled the 2 cpus
should i try to install macOSX ? here i have 10.3, should it help ?
i'd like to understand if it is an hw issue or a sw issue =P
p.s.
i have no way to take a screen picture, sorry =P |
|
Back to top |
|
|
blob999 n00b
Joined: 07 Jun 2009 Posts: 64 Location: NATO area
|
Posted: Tue Apr 26, 2011 11:48 am Post subject: |
|
|
well, i guess a kernel bug.
As i remember cruxppc bootkernel has smp builtin, give a try. _________________ LinuxPPC user!
my blog: http://linuxpowerpc.blogspot.com/ |
|
Back to top |
|
|
flameman n00b
Joined: 10 Apr 2011 Posts: 39
|
Posted: Tue Apr 26, 2011 12:25 pm Post subject: |
|
|
blob999 wrote: | well, i guess a kernel bug.
As i remember cruxppc bootkernel has smp builtin, give a try. |
exactly what do you meanby "cruxppc bootkernel" ?
ftp:/ /ftp.cruxppc.org/cruxppc/test/crux-ppc-2.7RC2.iso ?
i'd like to understand what kind of patched they apply in order to have a working SMP
i installed macOS10.3, which is pretty working and reporting 2x7450 cpus ... with no error in the dmesg |
|
Back to top |
|
|
blob999 n00b
Joined: 07 Jun 2009 Posts: 64 Location: NATO area
|
Posted: Tue Apr 26, 2011 12:31 pm Post subject: |
|
|
you need to use the latest stable (crux-ppc-2.7.iso).
You can download it from a mirror. _________________ LinuxPPC user!
my blog: http://linuxpowerpc.blogspot.com/ |
|
Back to top |
|
|
blob999 n00b
Joined: 07 Jun 2009 Posts: 64 Location: NATO area
|
Posted: Tue Apr 26, 2011 12:34 pm Post subject: |
|
|
flameman wrote: |
i'd like to understand what kind of patched they apply in order to have a working SMP
|
it isn't a patch question but it could be a bug in the linux kernel realease used on that specific gentoo install iso. _________________ LinuxPPC user!
my blog: http://linuxpowerpc.blogspot.com/ |
|
Back to top |
|
|
flameman n00b
Joined: 10 Apr 2011 Posts: 39
|
Posted: Tue Apr 26, 2011 3:43 pm Post subject: |
|
|
well, the crux does not boot at all
i noticed the cpu cooler was running too fast, so i did these 3 following things
1) I removed the big cpu thermal head shield in order to clean the cpu and to apply new thermal pad (metal sic)
2) I tried an other gentoo-ppc-mimimal-2010 bootcd (Dicember 2010)
3) I entered the open-firmware in order to hardly reset the PMU data
and i am now connected to the power mac
livecd ~ # cat /proc/cpuinfo
processor : 0
cpu : 7455, altivec supported
clock : 1249.999995MHz
revision : 3.2 (pvr 8001 0302)
bogomips : 83.32
processor : 1
cpu : 7455, altivec supported
clock : 1249.999995MHz
revision : 3.2 (pvr 8001 0302)
bogomips : 83.32
total bogomips : 166.64
timebase : 41660651
platform : PowerMac
model : PowerMac3,6
machine : PowerMac3,6
motherboard : PowerMac3,6 MacRISC2 MacRISC Power Macintosh
detected as : 129 (PowerMac G4 Windtunnel)
pmac flags : 00000010
L2 cache : 256K unified
pmac-generation : NewWorld
Memory : 512 MB
livecd ~ #
i noticed this time both the 2 cpu are correctly identified: removing the thermal head i read xc7455 written on both the 2 cpus, note macOSX was reporting a different type (7450 instead of 7455)
dunno which of these 3 things has solved, but now the machine is running gentoo (pretty downloading stage3 and portage) |
|
Back to top |
|
|
blob999 n00b
Joined: 07 Jun 2009 Posts: 64 Location: NATO area
|
Posted: Tue Apr 26, 2011 3:55 pm Post subject: |
|
|
glade you solved!
Anyway you'd understand if the issue was caused by running modern kernels because this does mean there is a serious bug in SMP subsystem for ppc32. _________________ LinuxPPC user!
my blog: http://linuxpowerpc.blogspot.com/ |
|
Back to top |
|
|
flameman n00b
Joined: 10 Apr 2011 Posts: 39
|
Posted: Tue Apr 26, 2011 6:28 pm Post subject: |
|
|
pretty test activity has already started
2.6.32-2.6.36 are under test |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|