Re: [XenPPC] Crash on js12
On Mon, Nov 27, 2006 at 12:04:36PM -0500, Maria Butrico wrote: > This problem is solved on the latest version of Xen. I do not think the problem is completely solved. I am still seeing repeated prints like this: (XEN) CPU[PIR:1 IPI:1 Logical:1] Hello World!CPU[PIR:1 IPI:1 Logical:1] Hello World! (XEN) :1] Hello World!Got ack ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] [xenppc-unstable] [XEN][POWERPC] SMP/IPI/MB combined
This will have to be reworked and broken up into individual parts for
submission, but here is what is necessary to make 'C-a C-a C-a t' work
properly.
Jimi, when you disassemble xen-syms compiled without this patch, do you
see a bogus infinite loop in read_clocks? The compiler is not told that
read_clocks_cpumask is volatile, so it is free to turn that loop into an
infinite loop, as my gcc 4.1.1 cross-compiler does. I am surprised that
other Xen architectures have not seen the same problem.
arch/powerpc/smp.c| 18 ++
common/keyhandler.c |2 +-
include/xen/cpumask.h |2 +-
3 files changed, 12 insertions(+), 10 deletions(-)
diff -r 305751a5281e xen/arch/powerpc/smp.c
--- a/xen/arch/powerpc/smp.cWed Nov 22 16:29:25 2006 -0500
+++ b/xen/arch/powerpc/smp.cTue Nov 28 00:45:24 2006 -0500
@@ -91,31 +91,33 @@ int on_selected_cpus(
int wait)
{
int t, retval = 0, nr_cpus = cpus_weight(selected);
+int stall = timebase_freq * 10;
spin_lock(&call_lock);
call_data.func = func;
call_data.info = info;
call_data.wait = wait;
-call_data.wait = 1; /* Until we get RCU around call_data. */
atomic_set(&call_data.started, 0);
atomic_set(&call_data.finished, 0);
mb();
send_IPI_mask(selected, CALL_FUNCTION_VECTOR);
-/* We always wait for an initiation ACK from remote CPU. */
-for (t = 0; atomic_read(&call_data.started) != nr_cpus; t++) {
-if (t && t % timebase_freq == 0) {
-printk("IPI start stall: %d ACKS to %d SYNS\n",
- atomic_read(&call_data.started), nr_cpus);
-}
+/* If told to, we wait for an initiation ACK from remote CPU. */
+if (wait) {
+ for (t = 0; atomic_read(&call_data.started) != nr_cpus; t++) {
+ if (t > 0 && t % stall == 0) {
+ printk("IPI start stall: %d ACKS to %d SYNS\n",
+ atomic_read(&call_data.started), nr_cpus);
+ }
+ }
}
/* If told to, we wait for a completion ACK from remote CPU. */
if (wait) {
for (t = 0; atomic_read(&call_data.finished) != nr_cpus; t++) {
-if (t > timebase_freq && t % timebase_freq == 0) {
+if (t > 0 && t % stall == 0) {
printk("IPI finish stall: %d ACKS to %d SYNS\n",
atomic_read(&call_data.finished), nr_cpus);
}
diff -r 305751a5281e xen/common/keyhandler.c
--- a/xen/common/keyhandler.c Wed Nov 22 16:29:25 2006 -0500
+++ b/xen/common/keyhandler.c Tue Nov 28 00:12:14 2006 -0500
@@ -193,7 +193,7 @@ static void dump_domains(unsigned char k
read_unlock(&domlist_lock);
}
-static cpumask_t read_clocks_cpumask = CPU_MASK_NONE;
+static cpumask_t volatile read_clocks_cpumask = CPU_MASK_NONE;
static s_time_t read_clocks_time[NR_CPUS];
static void read_clocks_slave(void *unused)
diff -r 305751a5281e xen/include/xen/cpumask.h
--- a/xen/include/xen/cpumask.h Wed Nov 22 16:29:25 2006 -0500
+++ b/xen/include/xen/cpumask.h Tue Nov 28 00:12:42 2006 -0500
@@ -177,7 +177,7 @@ static inline int __cpus_subset(const cp
}
#define cpus_empty(src) __cpus_empty(&(src), NR_CPUS)
-static inline int __cpus_empty(const cpumask_t *srcp, int nbits)
+static inline int __cpus_empty(const cpumask_t volatile *srcp, int nbits)
{
return bitmap_empty(srcp->bits, nbits);
}
___
Xen-ppc-devel mailing list
Xen-ppc-devel@lists.xensource.com
http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash on js12
On Mon, Nov 27, 2006 at 12:04:36PM -0500, Maria Butrico wrote: > This problem is solved on the latest version of Xen. I would encourage > Tony, who is running tests, to test all forms of xen, debug and not (I > am assuming that he uses an automated procedure, so is just a matter of > setting it up to go twice). xm-test wont help here as it doesn't boot dom0, all the tests are between dom0 and domU. Yours Tony linux.conf.au http://linux.conf.au/ || http://lca2007.linux.org.au/ Jan 15-20 2007 The Australian Linux Technical Conference! ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
[XenPPC] [xenppc-unstable] [XEN][POWERPC] OFD dump prefix screen and hook into keyhandler
# HG changeset patch
# User Jimi Xenidis <[EMAIL PROTECTED]>
# Node ID cc45282daf3d242fdcf6e858c0b18b7f1086a318
# Parent 02aaaf00a9229b6564e860618f329bd2f9303a1e
[XEN][POWERPC] OFD dump prefix screen and hook into keyhandler
This patch adds the ability to view the devtree from the Xen console.
Signed-off-by: Jimi Xenidis <[EMAIL PROTECTED]>
---
xen/arch/powerpc/boot_of.c|2 +-
xen/arch/powerpc/of-devtree.h |6 +++---
xen/arch/powerpc/of-devwalk.c | 14 +++---
xen/arch/powerpc/ofd_fixup.c |2 +-
xen/arch/powerpc/setup.c | 19 ++-
5 files changed, 26 insertions(+), 17 deletions(-)
diff -r 02aaaf00a922 -r cc45282daf3d xen/arch/powerpc/boot_of.c
--- a/xen/arch/powerpc/boot_of.cMon Nov 27 16:14:07 2006 -0500
+++ b/xen/arch/powerpc/boot_of.cMon Nov 27 17:17:07 2006 -0500
@@ -1069,7 +1069,7 @@ static void * __init boot_of_devtree(mod
if (ofd_size(oft) > oft_sz)
of_panic("Could not fit all devtree fixups\n");
-ofd_walk(oft, OFD_ROOT, /* add_hype_props */ NULL, 2);
+ofd_walk(oft, __func__, OFD_ROOT, /* add_hype_props */ NULL, 2);
mod->mod_start = (ulong)oft;
mod->mod_end = mod->mod_start + oft_sz;
diff -r 02aaaf00a922 -r cc45282daf3d xen/arch/powerpc/of-devtree.h
--- a/xen/arch/powerpc/of-devtree.h Mon Nov 27 16:14:07 2006 -0500
+++ b/xen/arch/powerpc/of-devtree.h Mon Nov 27 17:17:07 2006 -0500
@@ -115,10 +115,10 @@ extern void ofd_io_close(void *mem, ofdn
extern void ofd_io_close(void *mem, ofdn_t n);
-typedef void (*walk_fn)(void *m, ofdn_t p, int arg);
-extern void ofd_dump_props(void *m, ofdn_t p, int dump);
+typedef void (*walk_fn)(void *m, const char *pre, ofdn_t p, int arg);
+extern void ofd_dump_props(void *m, const char *pre, ofdn_t p, int dump);
-extern void ofd_walk(void *m, ofdn_t p, walk_fn fn, int arg);
+extern void ofd_walk(void *m, const char *pre, ofdn_t p, walk_fn fn, int arg);
/* Recursively look up #address_cells and #size_cells properties */
diff -r 02aaaf00a922 -r cc45282daf3d xen/arch/powerpc/of-devwalk.c
--- a/xen/arch/powerpc/of-devwalk.c Mon Nov 27 16:14:07 2006 -0500
+++ b/xen/arch/powerpc/of-devwalk.c Mon Nov 27 17:17:07 2006 -0500
@@ -80,7 +80,7 @@ void ofd_prop_print(
#endif
}
-void ofd_dump_props(void *mem, ofdn_t n, int dump)
+void ofd_dump_props(void *mem, const char *pre, ofdn_t n, int dump)
{
ofdn_t p;
char name[128];
@@ -95,7 +95,7 @@ void ofd_dump_props(void *mem, ofdn_t n,
}
if (dump & OFD_DUMP_NAMES) {
-printk("of_walk: %s: phandle 0x%x\n", path, n);
+printk("%s: %s: phandle 0x%x\n", pre, path, n);
}
p = ofd_nextprop(mem, n, NULL, name);
@@ -106,30 +106,30 @@ void ofd_dump_props(void *mem, ofdn_t n,
}
if ( dump & OFD_DUMP_VALUES ) {
-ofd_prop_print("of_walk", path, name, prop, sz);
+ofd_prop_print(pre, path, name, prop, sz);
}
p = ofd_nextprop(mem, n, name, name);
}
}
-void ofd_walk(void *m, ofdn_t p, walk_fn fn, int arg)
+void ofd_walk(void *m, const char *pre, ofdn_t p, walk_fn fn, int arg)
{
ofdn_t n;
if ( fn != NULL ) {
-(*fn)(m, p, arg);
+(*fn)(m, pre, p, arg);
}
/* child */
n = ofd_node_child(m, p);
if ( n != 0 ) {
-ofd_walk(m, n, fn, arg);
+ofd_walk(m, pre, n, fn, arg);
}
/* peer */
n = ofd_node_peer(m, p);
if ( n != 0 ) {
-ofd_walk(m, n, fn, arg);
+ofd_walk(m, pre, n, fn, arg);
}
}
diff -r 02aaaf00a922 -r cc45282daf3d xen/arch/powerpc/ofd_fixup.c
--- a/xen/arch/powerpc/ofd_fixup.c Mon Nov 27 16:14:07 2006 -0500
+++ b/xen/arch/powerpc/ofd_fixup.c Mon Nov 27 17:17:07 2006 -0500
@@ -427,7 +427,7 @@ int ofd_dom0_fixup(struct domain *d, ulo
#ifdef DEBUG
-ofd_walk(m, OFD_ROOT, ofd_dump_props, OFD_DUMP_ALL);
+ofd_walk(m, __func__, OFD_ROOT, ofd_dump_props, OFD_DUMP_ALL);
#endif
return 1;
}
diff -r 02aaaf00a922 -r cc45282daf3d xen/arch/powerpc/setup.c
--- a/xen/arch/powerpc/setup.c Mon Nov 27 16:14:07 2006 -0500
+++ b/xen/arch/powerpc/setup.c Mon Nov 27 17:17:07 2006 -0500
@@ -123,9 +123,17 @@ void noinline __attn(void)
console_end_sync();
}
-static void hw_probe_attn(unsigned char key, struct cpu_user_regs *regs)
+static void key_hw_probe_attn(unsigned char key)
{
__attn();
+}
+
+static void key_ofdump(unsigned char key)
+{
+printk("ofdump:\n");
+/* make sure the OF devtree is good */
+ofd_walk((void *)oftree, "devtree", OFD_ROOT,
+ ofd_dump_props, OFD_DUMP_ALL);
}
static void percpu_init_areas(void)
@@ -178,7 +186,10 @@ static void __init start_of_day(void)
/* Register another key that will allow for the the Harware Probe
* to be contacted, this works with RiscWatch probes and should
* work with Chronos and FSPs */
-register_irq_keyhandler('^', hw_probe_attn, "Trap to Hardware Probe");
+register_keyhandler('^
[XenPPC] [xenppc-unstable] [TOOLS][POWERPC] update prose builder to define the mem_mb arg.
# HG changeset patch
# User Jimi Xenidis <[EMAIL PROTECTED]>
# Node ID 02aaaf00a9229b6564e860618f329bd2f9303a1e
# Parent 305751a5281e12537a6485d74ac9f53a73ee8816
[TOOLS][POWERPC] update prose builder to define the mem_mb arg.
Signed-off-by: Jimi Xenidis <[EMAIL PROTECTED]>
---
tools/python/xen/xend/image.py |3 +++
1 files changed, 3 insertions(+)
diff -r 305751a5281e -r 02aaaf00a922 tools/python/xen/xend/image.py
--- a/tools/python/xen/xend/image.pyWed Nov 22 16:29:25 2006 -0500
+++ b/tools/python/xen/xend/image.pyMon Nov 27 16:14:07 2006 -0500
@@ -267,7 +267,10 @@ class PPC_ProseImageHandler(LinuxImageHa
store_evtchn = self.vm.getStorePort()
console_evtchn = self.vm.getConsolePort()
+mem_mb = self.getRequiredInitialReservation() / 1024
+
log.debug("dom= %d", self.vm.getDomid())
+log.debug("memsize= %d", mem_mb)
log.debug("image = %s", self.kernel)
log.debug("store_evtchn = %d", store_evtchn)
log.debug("console_evtchn = %d", console_evtchn)
___
Xen-ppc-devel mailing list
Xen-ppc-devel@lists.xensource.com
http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash with ssh
BTW: in the future, if you are lucky enough to get xmon, please do a
register dump ('r'), special register dump ('S'), and a function
traceback ('t') (use '?' for help) and send that as well.
-JX
On Nov 27, 2006, at 12:08 PM, Jimi Xenidis wrote:
On Nov 27, 2006, at 12:02 PM, Maria Butrico wrote:
Jimi, where did you get the idea that this linux was SMP. I
recall that we build Linux for SMP even though we run it UP if we
boot it on a xen partition.Maybe you see something I missed.
This:
cso85:/ # cpu 0x2: Vector: 700 (Program Check) at
[c0017b5a7340]
pc: c00f62bc: .clear_inode+0x3c/0x120
lr: c00f62b4: .clear_inode+0x34/0x120
sp: c0017b5a75c0
msr: 90029032
current = 0xc0002facc540
paca= 0xc05ae600
pid = 17947, comm = umount
kernel BUG in clear_inode at
/root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251!
enter ? for help
2:mon>
2:mon>
the BUG() output says "cpu 0x2:" and the # before the xmon prompt
also says CPU 2.
as you know we only give on VCPU to domains right now.
-JX
___
Xen-ppc-devel mailing list
Xen-ppc-devel@lists.xensource.com
http://lists.xensource.com/xen-ppc-devel
___
Xen-ppc-devel mailing list
Xen-ppc-devel@lists.xensource.com
http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash on js12
Thanks for verifying Maria. -JX On Nov 27, 2006, at 12:04 PM, Maria Butrico wrote: This problem is solved on the latest version of Xen. I would encourage Tony, who is running tests, to test all forms of xen, debug and not (I am assuming that he uses an automated procedure, so is just a matter of setting it up to go twice). Maria Butrico wrote: I seem to be the only person with this problem. On a js21 with 8G of memory, tip as of about 1 hour ago. (The machine can run xen of a couple of weeks ago).I have not tried on any other machine. OF: Xen/PPC version 3.0-unstable ([EMAIL PROTECTED]) (gcc version 4.1.1 ()) Mon Nov 20 15:44:04 EST 2006 boot_of_init args: 0x0 0x0 0x111027c 0x1275532 0x0 boot msr: 0x10003000 boot_of_init: _start 0040 _end 00493c78 0x1275532 boot_of_probemem: memory 0x[0x8000] boot_of_probemem: memory 0x0001[0x18000] bootargs = xen dom0_mem=2G -- console=xencons xencons=xvc0 ip=9.2.78.99::9.2.78.1:255.255.255.0:cso99:eth1:off ro root=/dev/ nfs nfsroot=9.2.208.204:/home/apw/devel/nfsroots/sles10- rootfs-2006-10-30 init=/sbin/quickinit noshell quick=/homes/kix/ butrico/libos/quickcfg-xen-ppc.sh boot_of_module: FYI Dom0 is unknown, will be caught later boot_of_module: dom0 mod @ 0x00493c78[0x493c78] boot_of_module: dom0 mod string: console=xencons xencons=xvc0 ip=9.2.78.99::9.2.78.1:255.255.255.0:cso99:eth1:off ro root=/dev/ nfs nfsroot=9.2.208.204:/home/apw/devel/nfsroots/sles10- rootfs-2006-10-30 init=/sbin/quickinit noshell quick=/homes/kix/ butrico/libos/quickcfg-xen-ppc.sh find_space base=0x00493c78 eomem=0x8000 size=0x8000 align=0x1000 find_space base=0x0049c000 eomem=0x8000 size=0x0003 align=0x1000 creating oftree pkg_save: saved device tree in 0x41c8 bytes boot_of_devtree: devtree mod @ 0x0049c000[0x4cc000] OF: timebase-frequency = 14318378 Hz OF: clock-frequency = 250 KHz ping = 0x: ping = 0x: ping = 0x: NOT FOUND get-property for device_type on zero phandle HANG ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash with ssh
On Nov 27, 2006, at 12:02 PM, Maria Butrico wrote: Jimi, where did you get the idea that this linux was SMP. I recall that we build Linux for SMP even though we run it UP if we boot it on a xen partition.Maybe you see something I missed. This: cso85:/ # cpu 0x2: Vector: 700 (Program Check) at [c0017b5a7340] pc: c00f62bc: .clear_inode+0x3c/0x120 lr: c00f62b4: .clear_inode+0x34/0x120 sp: c0017b5a75c0 msr: 90029032 current = 0xc0002facc540 paca= 0xc05ae600 pid = 17947, comm = umount kernel BUG in clear_inode at /root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! enter ? for help 2:mon> 2:mon> the BUG() output says "cpu 0x2:" and the # before the xmon prompt also says CPU 2. as you know we only give on VCPU to domains right now. -JX ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash on js12
This problem is solved on the latest version of Xen. I would encourage Tony, who is running tests, to test all forms of xen, debug and not (I am assuming that he uses an automated procedure, so is just a matter of setting it up to go twice). Maria Butrico wrote: I seem to be the only person with this problem. On a js21 with 8G of memory, tip as of about 1 hour ago. (The machine can run xen of a couple of weeks ago).I have not tried on any other machine. OF: Xen/PPC version 3.0-unstable ([EMAIL PROTECTED]) (gcc version 4.1.1 ()) Mon Nov 20 15:44:04 EST 2006 boot_of_init args: 0x0 0x0 0x111027c 0x1275532 0x0 boot msr: 0x10003000 boot_of_init: _start 0040 _end 00493c78 0x1275532 boot_of_probemem: memory 0x[0x8000] boot_of_probemem: memory 0x0001[0x18000] bootargs = xen dom0_mem=2G -- console=xencons xencons=xvc0 ip=9.2.78.99::9.2.78.1:255.255.255.0:cso99:eth1:off ro root=/dev/nfs nfsroot=9.2.208.204:/home/apw/devel/nfsroots/sles10-rootfs-2006-10-30 init=/sbin/quickinit noshell quick=/homes/kix/butrico/libos/quickcfg-xen-ppc.sh boot_of_module: FYI Dom0 is unknown, will be caught later boot_of_module: dom0 mod @ 0x00493c78[0x493c78] boot_of_module: dom0 mod string: console=xencons xencons=xvc0 ip=9.2.78.99::9.2.78.1:255.255.255.0:cso99:eth1:off ro root=/dev/nfs nfsroot=9.2.208.204:/home/apw/devel/nfsroots/sles10-rootfs-2006-10-30 init=/sbin/quickinit noshell quick=/homes/kix/butrico/libos/quickcfg-xen-ppc.sh find_space base=0x00493c78 eomem=0x8000 size=0x8000 align=0x1000 find_space base=0x0049c000 eomem=0x8000 size=0x0003 align=0x1000 creating oftree pkg_save: saved device tree in 0x41c8 bytes boot_of_devtree: devtree mod @ 0x0049c000[0x4cc000] OF: timebase-frequency = 14318378 Hz OF: clock-frequency = 250 KHz ping = 0x: ping = 0x: ping = 0x: NOT FOUND get-property for device_type on zero phandle HANG ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash with ssh
Jimi, where did you get the idea that this linux was SMP. I recall that we build Linux for SMP even though we run it UP if we boot it on a xen partition.Maybe you see something I missed. Jimi Xenidis wrote: Thanks for clarifying this Dilma, there is no Xen present if the Linux is SMP :) -JX On Nov 27, 2006, at 10:25 AM, Dilma DaSilva wrote: This is not a xen problem. I believe it's a gpfs problem on the shutdown/umount path. We're using gpfs with a kernel version that it doesn't support yet and I have fixed problems related to clear_inode before (Linux has changed some interfaces in the vfs layer on 2.6.17). Dilma David M Daly writes: I got the following error this morning from the console on one of our JS21s. I had a program that was supposed to ssh into the blade and reboot it after shutting down GPFS. I don't think it did any of the gpfs shutdown before crashing. Here is the error message. I saw it because I happened to be on the SOL console ( I wouldn't normally be). This is the first time I'm noticing this. I will post again if I see more errors of this kind. cso85:/ # cpu 0x2: Vector: 700 (Program Check) at [c0017b5a7340] pc: c00f62bc: .clear_inode+0x3c/0x120 lr: c00f62b4: .clear_inode+0x34/0x120 sp: c0017b5a75c0 msr: 90029032 current = 0xc0002facc540 paca= 0xc05ae600 pid = 17947, comm = umount kernel BUG in clear_inode at /root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! enter ? for help 2:mon> 2:mon> This is built from Maria's tree on 11-9. David Daly Research Staff Member 32-003 IBM T.J. Watson Research Center Yorktown Heights, NY, 10958 (914) 945-1845, T/L 862-1845 I got the following error this morning from the console on one of our JS21s. I had a program that was supposed to ssh into the blade and reboot it after shutting down GPFS. I don't think it did any of the gpfs shutdown before crashing. Here is the error message. I saw it because I happened to be on the SOL console ( I wouldn't normally be). This is the first time I'm noticing this. I will post again if I see more errors of this kind. cso85:/ # cpu 0x2: Vector: 700 (Program Check) at [c0017b5a7340] pc: c00f62bc: .clear_inode+0x3c/0x120 lr: c00f62b4: .clear_inode+0x34/0x120 sp: c0017b5a75c0 msr: 90029032 current = 0xc0002facc540 paca = 0xc05ae600 pid = 17947, comm = umount kernel BUG in clear_inode at /root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! enter ? for help 2:mon> 2:mon> This is built from Maria's tree on 11-9. David Daly Research Staff Member 32-003 IBM T.J. Watson Research Center Yorktown Heights, NY, 10958 (914) 945-1845, T/L 862-1845___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash with ssh
Thanks for clarifying this Dilma, there is no Xen present if the Linux is SMP :) -JX On Nov 27, 2006, at 10:25 AM, Dilma DaSilva wrote: This is not a xen problem. I believe it's a gpfs problem on the shutdown/umount path. We're using gpfs with a kernel version that it doesn't support yet and I have fixed problems related to clear_inode before (Linux has changed some interfaces in the vfs layer on 2.6.17). Dilma David M Daly writes: I got the following error this morning from the console on one of our JS21s. I had a program that was supposed to ssh into the blade and reboot it after shutting down GPFS. I don't think it did any of the gpfs shutdown before crashing. Here is the error message. I saw it because I happened to be on the SOL console ( I wouldn't normally be). This is the first time I'm noticing this. I will post again if I see more errors of this kind. cso85:/ # cpu 0x2: Vector: 700 (Program Check) at [c0017b5a7340] pc: c00f62bc: .clear_inode+0x3c/0x120 lr: c00f62b4: .clear_inode+0x34/0x120 sp: c0017b5a75c0 msr: 90029032 current = 0xc0002facc540 paca= 0xc05ae600 pid = 17947, comm = umount kernel BUG in clear_inode at /root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! enter ? for help 2:mon> 2:mon> This is built from Maria's tree on 11-9. David Daly Research Staff Member 32-003 IBM T.J. Watson Research Center Yorktown Heights, NY, 10958 (914) 945-1845, T/L 862-1845 I got the following error this morning from the console on one of our JS21s. I had a program that was supposed to ssh into the blade and reboot it after shutting down GPFS. I don't think it did any of the gpfs shutdown before crashing. Here is the error message. I saw it because I happened to be on the SOL console ( I wouldn't normally be). This is the first time I'm noticing this. I will post again if I see more errors of this kind. cso85:/ # cpu 0x2: Vector: 700 (Program Check) at [c0017b5a7340] pc: c00f62bc: .clear_inode+0x3c/0x120 lr: c00f62b4: .clear_inode+0x34/0x120 sp: c0017b5a75c0 msr: 90029032 current = 0xc0002facc540 paca = 0xc05ae600 pid = 17947, comm = umount kernel BUG in clear_inode at / root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! enter ? for help 2:mon> 2:mon> This is built from Maria's tree on 11-9. David Daly Research Staff Member 32-003 IBM T.J. Watson Research Center Yorktown Heights, NY, 10958 (914) 945-1845, T/L 862-1845font>___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
Re: [XenPPC] Crash with ssh
This is not a xen problem. I believe it's a gpfs problem on the shutdown/umount path. We're using gpfs with a kernel version that it doesn't support yet and I have fixed problems related to clear_inode before (Linux has changed some interfaces in the vfs layer on 2.6.17). Dilma David M Daly writes: > I got the following error this morning from the console on one of our > JS21s. I had a program that was supposed to ssh into the blade and reboot > it after shutting down GPFS. I don't think it did any of the gpfs shutdown > before crashing. Here is the error message. I saw it because I happened to > be on the SOL console ( I wouldn't normally be). This is the first time > I'm noticing this. I will post again if I see more errors of this kind. > > cso85:/ # cpu 0x2: Vector: 700 (Program Check) at [c0017b5a7340] > pc: c00f62bc: .clear_inode+0x3c/0x120 > lr: c00f62b4: .clear_inode+0x34/0x120 > sp: c0017b5a75c0 >msr: 90029032 > current = 0xc0002facc540 > paca= 0xc05ae600 > pid = 17947, comm = umount > kernel BUG in clear_inode at > /root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! > enter ? for help > 2:mon> > 2:mon> > > This is built from Maria's tree on 11-9. > > David Daly > Research Staff Member > 32-003 IBM T.J. Watson Research Center > Yorktown Heights, NY, 10958 > (914) 945-1845, T/L 862-1845 > I got the following error this morning > from the console on one of our JS21s. I had a program that was supposed > to ssh into the blade and reboot it after shutting down GPFS. I don't think > it did any of the gpfs shutdown before crashing. Here is the error message. > I saw it because I happened to be on the SOL console ( I wouldn't normally > be). This is the first time I'm noticing this. I will post again if I see > more errors of this kind. > > cso85:/ # cpu 0x2: Vector: 700 (Program > Check) at [c0017b5a7340] > pc: c00f62bc: > .clear_inode+0x3c/0x120 > lr: c00f62b4: > .clear_inode+0x34/0x120 > sp: c0017b5a75c0 > msr: 90029032 > current = 0xc0002facc540 > paca = > 0xc05ae600 > pid = 17947, comm > = umount > kernel BUG in clear_inode at > /root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! > enter ? for help > 2:mon> > 2:mon> > > This is built from Maria's tree on 11-9. > > > David Daly > Research Staff Member > 32-003 IBM T.J. Watson Research Center > Yorktown Heights, NY, 10958 > (914) 945-1845, T/L > 862-1845___ > Xen-ppc-devel mailing list > Xen-ppc-devel@lists.xensource.com > http://lists.xensource.com/xen-ppc-devel ___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
[XenPPC] Crash with ssh
I got the following error this morning from the console on one of our JS21s. I had a program that was supposed to ssh into the blade and reboot it after shutting down GPFS. I don't think it did any of the gpfs shutdown before crashing. Here is the error message. I saw it because I happened to be on the SOL console ( I wouldn't normally be). This is the first time I'm noticing this. I will post again if I see more errors of this kind. cso85:/ # cpu 0x2: Vector: 700 (Program Check) at [c0017b5a7340] pc: c00f62bc: .clear_inode+0x3c/0x120 lr: c00f62b4: .clear_inode+0x34/0x120 sp: c0017b5a75c0 msr: 90029032 current = 0xc0002facc540 paca= 0xc05ae600 pid = 17947, comm = umount kernel BUG in clear_inode at /root/xen-maria-latest/linux/linux-xen-ppc/fs/inode.c:251! enter ? for help 2:mon> 2:mon> This is built from Maria's tree on 11-9. David Daly Research Staff Member 32-003 IBM T.J. Watson Research Center Yorktown Heights, NY, 10958 (914) 945-1845, T/L 862-1845___ Xen-ppc-devel mailing list Xen-ppc-devel@lists.xensource.com http://lists.xensource.com/xen-ppc-devel
