rcu: Fix CONFIG_RCU_NOCB_CPU_ALL panic on machines with sparse CPU mask
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 34 lines · Kirill Tkhai
When a system has a sparse cpumask and
CONFIG_RCU_NOCB_CPU_ALLis enabled,rcu_spawn_nocb_kthreads()creates nocb threads for nonexistent CPUS.The problem is in
rcu_bootup_announce_oddness().cpumask_setall()sets all bits from 0 tonr_cpu_idsexistent and nonexistent CPUs both.The sample. My cpuinfo(sparc64):
CPU0: online
CPU2: onlineI get these oopses every boot:
Unable to handle kernel NULL pointer dereference CPU: 0 PID: 23 Comm: rcuos/1 Not tainted 3.11.0+ #103 task: fffff800be9792c0 ti: fffff800be98c000 task.ti: fffff800be98c000 TSTATE: 0000004480e01604 TPC: 000000000047cffc TNPC: 000000000047d000 Y: 00000000 Not tainted TPC: <prepare_to_wait+0x38/0x5c> g0: 0000000000000000 g1: 0000000000000000 g2: fffff800be98fd58 g3: 0000000000acb718 g4: fffff800be9792c0 g5: fffff800be536000 g6: fffff800be98c000 g7: 0000000000000000 o0: 0000000000000000 o1: 0000000000a33d34 o2: 0000000000000001 o3: 0000000000000000 o4: 0000000000000000 o5: 0000000000000000 sp: fffff800be98f3d1 ret_pc: 000000000047cfd8 RPC: <prepare_to_wait+0x14/0x5c> l0: 0000000000a33ce8 l1: 0000000000000001 l2: 0000000000a7d800 l3: fffff800bf004dc0 l4: fffff800be880310 l5: 0000000000a33ce8 l6: 0000000000ad0000 l7: 0000000000a33800 i0: 0000000000acb710 i1: fffff800be98fd40 i2: 0000000000000001 i3: fffff800be98fd40 i4: 0000000000000000 i5: 0000000000000000 i6: fffff800be98f491 i7: 00000000004d3f50 I7: <rcu_nocb_kthread+0x5c/0x224> Call Trace: [00000000004d3f50] rcu_nocb_kthread+0x5c/0x224 [000000000047c604] kthread+0x88/0x9c [0000000000406084] ret_from_fork+0x1c/0x2c [0000000000000000] (null)
Quoted 1 line
So, not possible CPUs have to be cleared.
Good catch!
I am hoping that your setup always has CPU 0 defined. If not, there are many other problems in addition to CONFIG_RCU_NOCB_CPU_ZERO=y!
Quoted 11 lines
Signed-off-by Kirill Tkhai <[email protected]>CC Dipankar Sarma <[email protected]>CC Paul E. McKenney <[email protected]>--- kernel/rcutree_plugin.h | 1 + 1 files changed, 1 insertions(+), 0 deletions(-)kernel/rcutree_plugin.h +0
-0
kernel/rcutree_plugin.h static void __init rcu_bootup_announce_oddness(void) · line 100
Quoted 4 lines
kernel/rcutree_plugin.h +1
-0@@ -97,6 +97,7 @@ static void __init rcu_bootup_announce_oddness(void)#ifdef CONFIG_RCU_NOCB_CPU_ALL pr_info ("\tOffload RCU callbacks from all CPUs\n" );cpumask_setall (rcu_nocb_mask); +cpumask_and (rcu_nocb_mask, cpu_possible_mask, rcu_nocb_mask);
Would it make sense to combine the cpumask_setall() and the cpumask_and() into a single cpumask_copy()?
kernel/rcutree_plugin.h static void __init rcu_bootup_announce_oddness(void) · line 103
Quoted 3 lines
kernel/rcutree_plugin.h +0
-0#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */ #endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */ if (have_rcu_nocb_mask) {
Do we need a cpumask_and() within this "if" statement to handle the case where the kernel's rcu_nocbs= boot parameter specified a CPU that is not present? (If so, probably also giving a warning about the non-present CPUs specified.)
I believe that the answer to both questions is "yes", and I would welcome an updated patch that covered these cases.
All that said... Why does the system have a sparse cpu_possible_mask? This does cause overallocation of memory in a number of areas, including RCU's combining tree of rcu_node structures, to say nothing of the cpumasks themselves.
Thanx, Paul
14.09.2013, 22:48, "Paul E. McKenney" <[email protected]>:
Quoted 36 lines · Kirill Tkhai
Quoted 35 lines
When a system has a sparse cpumask and
CONFIG_RCU_NOCB_CPU_ALLis enabled,rcu_spawn_nocb_kthreads()creates nocb threads for nonexistent CPUS.The problem is in
rcu_bootup_announce_oddness().cpumask_setall()sets all bits from 0 tonr_cpu_idsexistent and nonexistent CPUs both.The sample. My cpuinfo(sparc64):
CPU0: online
CPU2: onlineI get these oopses every boot:
Unable to handle kernel NULL pointer dereference CPU: 0 PID: 23 Comm: rcuos/1 Not tainted 3.11.0+ #103 task: fffff800be9792c0 ti: fffff800be98c000 task.ti: fffff800be98c000 TSTATE: 0000004480e01604 TPC: 000000000047cffc TNPC: 000000000047d000 Y: 00000000 Not tainted TPC: <prepare_to_wait+0x38/0x5c> g0: 0000000000000000 g1: 0000000000000000 g2: fffff800be98fd58 g3: 0000000000acb718 g4: fffff800be9792c0 g5: fffff800be536000 g6: fffff800be98c000 g7: 0000000000000000 o0: 0000000000000000 o1: 0000000000a33d34 o2: 0000000000000001 o3: 0000000000000000 o4: 0000000000000000 o5: 0000000000000000 sp: fffff800be98f3d1 ret_pc: 000000000047cfd8 RPC: <prepare_to_wait+0x14/0x5c> l0: 0000000000a33ce8 l1: 0000000000000001 l2: 0000000000a7d800 l3: fffff800bf004dc0 l4: fffff800be880310 l5: 0000000000a33ce8 l6: 0000000000ad0000 l7: 0000000000a33800 i0: 0000000000acb710 i1: fffff800be98fd40 i2: 0000000000000001 i3: fffff800be98fd40 i4: 0000000000000000 i5: 0000000000000000 i6: fffff800be98f491 i7: 00000000004d3f50 I7: <rcu_nocb_kthread+0x5c/0x224> Call Trace: [00000000004d3f50] rcu_nocb_kthread+0x5c/0x224 [000000000047c604] kthread+0x88/0x9c [0000000000406084] ret_from_fork+0x1c/0x2c[0000000000000000] (null)
Quoted 6 lines
Quoted 1 line
So, not possible CPUs have to be cleared.
Good catch!
I am hoping that your setup always has CPU 0 defined. If not, there are many other problems in addition to
CONFIG_RCU_NOCB_CPU_ZERO=y!
I've never seen machines like you said, but it seem they are possible. boot_cpu_init() accounts the fact kernel may boot from not zero CPU:
static void __init boot_cpu_init(void)
{
int cpu = smp_processor_id();
/* Mark the boot cpu "present", "online" etc for SMP and UP case */
set_cpu_online(cpu, true);
set_cpu_active(cpu, true);
set_cpu_present(cpu, true);
set_cpu_possible(cpu, true);
}Quoted 23 lines
Quoted 6 lines
Signed-off-by Kirill Tkhai <[email protected]>CC Dipankar Sarma <[email protected]>CC Paul E. McKenney <[email protected]>---
kernel/rcutree_plugin.h | 1 +1 files changed, 1 insertions(+), 0 deletions(-)Quoted 9 lines
diff --git
a/kernel/rcutree_plugin.hb/kernel/rcutree_plugin.hindex 130c97b..1f17e99 100644
---a/kernel/rcutree_plugin.h
+++b/kernel/rcutree_plugin.h
@@ -97,6 +97,7 @@ static void__initrcu_bootup_announce_oddness(void)#ifdefCONFIG_RCU_NOCB_CPU_ALL
pr_info("\tOffload RCU callbacks from all CPUs\n");cpumask_setall(rcu_nocb_mask);
+cpumask_and(rcu_nocb_mask, cpu_possible_mask, rcu_nocb_mask);Would it make sense to combine the
cpumask_setall()and thecpumask_and()into a singlecpumask_copy()?Quoted 3 lines
#endif /* #ifdef
CONFIG_RCU_NOCB_CPU_ALL*/
#endif /* #ifndefCONFIG_RCU_NOCB_CPU_NONE*/
if (have_rcu_nocb_mask) {
Quoted 7 lines
Do we need a
cpumask_and()within this "if" statement to handle the case where the kernel's rcu_nocbs= boot parameter specified a CPU that is not present? (If so, probably also giving a warning about the non-present CPUs specified.)I believe that the answer to both questions is "yes", and I would welcome an updated patch that covered these cases.
Yes and yes. New patch is below.
Quoted 4 lines
All that said... Why does the system have a sparse
cpu_possible_mask? This does cause overallocation of memory in a number of areas, including RCU's combining tree ofrcu_nodestructures, to say nothing of the cpumasks themselves.
In case of sparc64 the cpu number is configured by boot/openprom. Kernel doesn't change it. I have not touched inside of them, so I don't know.
[PATCH] rcu: Fix CONFIG_RCU_NOCB_CPU_ALL panic on machines with sparse CPU mask
Some architectures have sparse cpu mask. UltraSparc's cpuinfo for example:
CPU0: online
CPU2: online
So, set only possible CPUs when CONFIG_RCU_NOCB_CPU_ALL is enabled.
Also, check that user passes right 'rcu_nocbs=' option.
--- kernel/rcutree_plugin.h | 8 +++++++- 1 files changed, 7 insertions(+), 1 deletions(-)
kernel/rcutree_plugin.h +7 -1
@@ -96,10 +96,16 @@ static void __init rcu_bootup_announce_oddness(void)#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ZERO */ #ifdef CONFIG_RCU_NOCB_CPU_ALL pr_info ("\tOffload RCU callbacks from all CPUs\n" ); -cpumask_setall (rcu_nocb_mask); +cpumask_copy (rcu_nocb_mask, cpu_possible_mask);#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */ #endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */ if (have_rcu_nocb_mask) { +if (!cpumask_subset (rcu_nocb_mask, cpu_possible_mask)) { +pr_info ("\tNote: kernel parameter 'rcu_nocbs=' " +"contains nonexistent CPUs.\n" ); +cpumask_and (rcu_nocb_mask, cpu_possible_mask, + rcu_nocb_mask); + }cpulist_scnprintf (nocb_buf,sizeof (nocb_buf), rcu_nocb_mask);pr_info ("\tOffload RCU callbacks from CPUs: %s.\n" , nocb_buf);if (rcu_nocb_poll)
Quoted 105 lines · Kirill Tkhai
14.09.2013, 22:48, "Paul E. McKenney" <[email protected]>:
Quoted 36 lines · Kirill Tkhai
Quoted 35 lines
When a system has a sparse cpumask and
CONFIG_RCU_NOCB_CPU_ALLis enabled,rcu_spawn_nocb_kthreads()creates nocb threads for nonexistent CPUS.The problem is in
rcu_bootup_announce_oddness().cpumask_setall()sets all bits from 0 tonr_cpu_idsexistent and nonexistent CPUs both.The sample. My cpuinfo(sparc64):
CPU0: online
CPU2: onlineI get these oopses every boot:
Unable to handle kernel NULL pointer dereference CPU: 0 PID: 23 Comm: rcuos/1 Not tainted 3.11.0+ #103 task: fffff800be9792c0 ti: fffff800be98c000 task.ti: fffff800be98c000 TSTATE: 0000004480e01604 TPC: 000000000047cffc TNPC: 000000000047d000 Y: 00000000 Not tainted TPC: <prepare_to_wait+0x38/0x5c> g0: 0000000000000000 g1: 0000000000000000 g2: fffff800be98fd58 g3: 0000000000acb718 g4: fffff800be9792c0 g5: fffff800be536000 g6: fffff800be98c000 g7: 0000000000000000 o0: 0000000000000000 o1: 0000000000a33d34 o2: 0000000000000001 o3: 0000000000000000 o4: 0000000000000000 o5: 0000000000000000 sp: fffff800be98f3d1 ret_pc: 000000000047cfd8 RPC: <prepare_to_wait+0x14/0x5c> l0: 0000000000a33ce8 l1: 0000000000000001 l2: 0000000000a7d800 l3: fffff800bf004dc0 l4: fffff800be880310 l5: 0000000000a33ce8 l6: 0000000000ad0000 l7: 0000000000a33800 i0: 0000000000acb710 i1: fffff800be98fd40 i2: 0000000000000001 i3: fffff800be98fd40 i4: 0000000000000000 i5: 0000000000000000 i6: fffff800be98f491 i7: 00000000004d3f50 I7: <rcu_nocb_kthread+0x5c/0x224> Call Trace: [00000000004d3f50] rcu_nocb_kthread+0x5c/0x224 [000000000047c604] kthread+0x88/0x9c [0000000000406084] ret_from_fork+0x1c/0x2c[0000000000000000] (null)
Quoted 6 lines
Quoted 1 line
So, not possible CPUs have to be cleared.
Good catch!
I am hoping that your setup always has CPU 0 defined. If not, there are many other problems in addition to
CONFIG_RCU_NOCB_CPU_ZERO=y!I've never seen machines like you said, but it seem they are possible.
boot_cpu_init()accounts the fact kernel may boot from not zero CPU:static void
__initboot_cpu_init(void){ int cpu = smp_processor_id(); /* Mark the boot cpu "present", "online" etc for SMP and UP case */ set_cpu_online(cpu, true); set_cpu_active(cpu, true); set_cpu_present(cpu, true); set_cpu_possible(cpu, true); }Quoted 23 lines
Quoted 6 lines
Signed-off-by Kirill Tkhai <[email protected]>CC Dipankar Sarma <[email protected]>CC Paul E. McKenney <[email protected]>---
kernel/rcutree_plugin.h | 1 +1 files changed, 1 insertions(+), 0 deletions(-)Quoted 9 lines
diff --git
a/kernel/rcutree_plugin.hb/kernel/rcutree_plugin.hindex 130c97b..1f17e99 100644
---a/kernel/rcutree_plugin.h
+++b/kernel/rcutree_plugin.h
@@ -97,6 +97,7 @@ static void__initrcu_bootup_announce_oddness(void)#ifdefCONFIG_RCU_NOCB_CPU_ALL
pr_info("\tOffload RCU callbacks from all CPUs\n");cpumask_setall(rcu_nocb_mask);
+cpumask_and(rcu_nocb_mask, cpu_possible_mask, rcu_nocb_mask);Would it make sense to combine the
cpumask_setall()and thecpumask_and()into a singlecpumask_copy()?Quoted 3 lines
#endif /* #ifdef
CONFIG_RCU_NOCB_CPU_ALL*/
#endif /* #ifndefCONFIG_RCU_NOCB_CPU_NONE*/
if (have_rcu_nocb_mask) {Quoted 7 lines
Do we need a
cpumask_and()within this "if" statement to handle the case where the kernel's rcu_nocbs= boot parameter specified a CPU that is not present? (If so, probably also giving a warning about the non-present CPUs specified.)I believe that the answer to both questions is "yes", and I would welcome an updated patch that covered these cases.
Yes and yes. New patch is below.
Quoted 4 lines
All that said... Why does the system have a sparse
cpu_possible_mask? This does cause overallocation of memory in a number of areas, including RCU's combining tree ofrcu_nodestructures, to say nothing of the cpumasks themselves.In case of sparc64 the cpu number is configured by boot/openprom. Kernel doesn't change it. I have not touched inside of them, so I don't know.
[PATCH] rcu: Fix
CONFIG_RCU_NOCB_CPU_ALLpanic on machines with sparse CPU maskSome architectures have sparse cpu mask. UltraSparc's cpuinfo for example:
CPU0: online
CPU2: online
Quoted 7 lines
So, set only possible CPUs when
CONFIG_RCU_NOCB_CPU_ALLis enabled.Also, check that user passes right 'rcu_nocbs=' option.
Signed-off-by Kirill Tkhai <[email protected]>CC Dipankar Sarma <[email protected]>CC Paul E. McKenney <[email protected]>
Queued for 3.13 with modification called out by scripts/checkpatch.pl. Please run scripts/checkpatch.pl on future patches before submitting.
Thanx, Paul
Quoted 26 lines
--- kernel/rcutree_plugin.h | 8 +++++++- 1 files changed, 7 insertions(+), 1 deletions(-)kernel/rcutree_plugin.h +7
-1@@ -96,10 +96,16 @@ static void __init rcu_bootup_announce_oddness(void)#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ZERO */ #ifdef CONFIG_RCU_NOCB_CPU_ALL pr_info ("\tOffload RCU callbacks from all CPUs\n" ); -cpumask_setall (rcu_nocb_mask); +cpumask_copy (rcu_nocb_mask, cpu_possible_mask);#endif /* #ifdef CONFIG_RCU_NOCB_CPU_ALL */ #endif /* #ifndef CONFIG_RCU_NOCB_CPU_NONE */ if (have_rcu_nocb_mask) { +if (!cpumask_subset (rcu_nocb_mask, cpu_possible_mask)) { +pr_info ("\tNote: kernel parameter 'rcu_nocbs=' " +"contains nonexistent CPUs.\n" ); +cpumask_and (rcu_nocb_mask, cpu_possible_mask, + rcu_nocb_mask); + }cpulist_scnprintf (nocb_buf,sizeof (nocb_buf), rcu_nocb_mask);pr_info ("\tOffload RCU callbacks from CPUs: %s.\n" , nocb_buf);if (rcu_nocb_poll)