An ssp initialized before rcu_init_geometry() will have its snp hierarchy
based on CONFIG_NR_CPUS.
Once rcu_init_geometry() is called, the nodes distribution is shrinked
and optimized toward meeting the actual possible number of CPUs detected
on boot.
Later on, the ssp that was initialized before rcu_init_geometry() is
confused and sometimes refers to its initial CONFIG_NR_CPUS based node
hierarchy, sometimes to the new num_possible_cpus() based one instead.
For example each of its sdp->mynode remain backward and refer to the
early node leaves that may not exist anymore. On the other hand the
srcu_for_each_node_breadth_first() refers to the new node hierarchy.
There are at least two bad possible outcomes to this:
1) a) A callback enqueued early on an sdp is recorded pending on
sdp->mynode->srcu_data_have_cbs in srcu_funnel_gp_start() with
sdp->mynode pointing to a deep leaf (say 3 levels).
b) The grace period ends after rcu_init_geometry() which shrinks the
nodes level to a single one. srcu_gp_end() walks through the new
snp hierarchy without ever reaching the old leaves so the callback
is never executed.
This is easily reproduced on an 8 CPUs machine with
CONFIG_NR_CPUS >= 32 and "rcupdate.rcu_self_test=1". The
srcu_barrier() after early tests verification never completes and
the boot hangs:
[ 5413.141029] INFO: task swapper/0:1 blocked for more than 4915
seconds.
[ 5413.147564] Not tainted 5.12.0-rc4+ #28
[ 5413.151927] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
disables this message.
[ 5413.159753] task:swapper/0 state:D stack: 0 pid: 1 ppid:
0 flags:0x00004000
[ 5413.168099] Call Trace:
[ 5413.170555] __schedule+0x36c/0x930
[ 5413.174057] ? wait_for_completion+0x88/0x110
[ 5413.178423] schedule+0x46/0xf0
[ 5413.181575] schedule_timeout+0x284/0x380
[ 5413.185591] ? wait_for_completion+0x88/0x110
[ 5413.189957] ? mark_held_locks+0x61/0x80
[ 5413.193882] ? mark_held_locks+0x61/0x80
[ 5413.197809] ? _raw_spin_unlock_irq+0x24/0x50
[ 5413.202173] ? wait_for_completion+0x88/0x110
[ 5413.206535] wait_for_completion+0xb4/0x110
[ 5413.210724] ? srcu_torture_stats_print+0x110/0x110
[ 5413.215610] srcu_barrier+0x187/0x200
[ 5413.219277] ? rcu_tasks_verify_self_tests+0x50/0x50
[ 5413.224244] ? rdinit_setup+0x2b/0x2b
[ 5413.227907] rcu_verify_early_boot_tests+0x2d/0x40
[ 5413.232700] do_one_initcall+0x63/0x310
[ 5413.236541] ? rdinit_setup+0x2b/0x2b
[ 5413.240207] ? rcu_read_lock_sched_held+0x52/0x80
[ 5413.244912] kernel_init_freeable+0x253/0x28f
[ 5413.249273] ? rest_init+0x250/0x250
[ 5413.252846] kernel_init+0xa/0x110
[ 5413.256257] ret_from_fork+0x22/0x30
2) An ssp that gets initialized before rcu_init_geometry() and used
afterward will always have stale rdp->mynode references, resulting in
callbacks to be missed in srcu_gp_end(), just like in the previous
scenario.
Solve this once rcu_init_geometry() is done with resetting the whole
state and node tree layout for each early initialized ssp's. Queued
callbacks are saved then requeued once the ssp reset is done.
Suggested-by: Paul E . McKenney <paul...@kernel.org>
Signed-off-by: Frederic Weisbecker <frede...@kernel.org>
Cc: Boqun Feng <boqun.f...@gmail.com>
Cc: Lai Jiangshan <jiangshan...@gmail.com>
Cc: Neeraj Upadhyay <neer...@codeaurora.org>
Cc: Josh Triplett <j...@joshtriplett.org>
Cc: Joel Fernandes <j...@joelfernandes.org>
Cc: Uladzislau Rezki <ure...@gmail.com>
---
include/linux/srcutree.h | 1 +
kernel/rcu/srcutree.c | 77 ++++++++++++++++++++++++++++++++++------
2 files changed, 68 insertions(+), 10 deletions(-)
diff --git a/include/linux/srcutree.h b/include/linux/srcutree.h
index cb1f4351e8ba..a2422c442470 100644
--- a/include/linux/srcutree.h
+++ b/include/linux/srcutree.h
@@ -83,6 +83,7 @@ struct srcu_struct {
/* operation. */
struct delayed_work work;
struct lockdep_map dep_map;
+ struct list_head early_init;
};
/* Values for state variable (bottom bits of ->srcu_gp_seq). */
diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c
index 108f9ca06047..7ca1bd0067c4 100644
--- a/kernel/rcu/srcutree.c
+++ b/kernel/rcu/srcutree.c
@@ -39,7 +39,7 @@ static ulong counter_wrap_check = (ULONG_MAX >> 2);
module_param(counter_wrap_check, ulong, 0444);
/* Early-boot callback-management, so early that no lock is required! */
-static LIST_HEAD(srcu_boot_list);
+static LIST_HEAD(srcu_early_init_list);
static bool __read_mostly srcu_init_done;
static void srcu_invoke_callbacks(struct work_struct *work);
@@ -174,10 +174,61 @@ static int init_srcu_struct_fields(struct srcu_struct
*ssp, bool is_static)
init_srcu_struct_nodes(ssp);
ssp->srcu_gp_seq_needed_exp = 0;
ssp->srcu_last_gp_end = ktime_get_mono_fast_ns();
+ if (!srcu_init_done)
+ list_add_tail(&ssp->early_init, &srcu_early_init_list);
smp_store_release(&ssp->srcu_gp_seq_needed, 0); /* Init done. */
return 0;
}
+/*
+ * After rcu_init_geometry(), early initialized SRCU nodes can be
+ * in a confused state, with sdp->mynode pointing to pruned nodes.
+ * Reset the ssp struct along with its node hierarchy and requeue its
+ * callbacks.
+ */
+static void reset_srcu_struct(struct srcu_struct *ssp)
+{
+ int cpu;
+ struct lockdep_map dep_map;
+ struct rcu_cblist pendcbs;
+ struct rcu_head *rhp;
+ struct srcu_data __percpu *sda;
+
+ rcu_cblist_init(&pendcbs);
+
+ for_each_possible_cpu(cpu) {
+ struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu);
+
+ if (cpu == smp_processor_id())
+ rcu_segcblist_extract_pend_cbs(&sdp->srcu_cblist,
&pendcbs);
+
+ /*
+ * We don't expect to have ready callbacks at this stage, nor
any
+ * callback at all on non-boot CPUs.
+ */
+ WARN_ON_ONCE(!rcu_segcblist_empty(&sdp->srcu_cblist));
+
+ memset(sdp, 0, sizeof(*sdp));
+ }
+
+ /*
+ * Save the per-CPU states. We don't need to re-allocate them.
+ * Besides they may even be static.
+ */
+ sda = ssp->sda;
+ /* Save the lockdep map, it may not suffer double-initialization */
+ dep_map = ssp->dep_map;
+
+ memset(ssp, 0, sizeof(*ssp));
+ ssp->sda = sda;
+ ssp->dep_map = dep_map;
+ spin_lock_init(&ACCESS_PRIVATE(ssp, lock));
+ init_srcu_struct_fields(ssp, true);
+
+ while ((rhp = rcu_cblist_dequeue(&pendcbs)))
+ call_srcu(ssp, rhp, rhp->func);
+}
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
int __init_srcu_struct(struct srcu_struct *ssp, const char *name,
@@ -675,11 +726,14 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp,
struct srcu_data *sdp,
rcu_seq_state(ssp->srcu_gp_seq) == SRCU_STATE_IDLE) {
WARN_ON_ONCE(ULONG_CMP_GE(ssp->srcu_gp_seq,
ssp->srcu_gp_seq_needed));
srcu_gp_start(ssp);
+ /*
+ * If SRCU isn't initialized yet, the grace period will be
+ * started again on srcu_init() at which point the work will
+ * be safely queued to rcu_gp_wq.
+ */
if (likely(srcu_init_done))
queue_delayed_work(rcu_gp_wq, &ssp->work,
srcu_get_delay(ssp));
- else if (list_empty(&ssp->work.work.entry))
- list_add(&ssp->work.work.entry, &srcu_boot_list);
}
spin_unlock_irqrestore_rcu_node(ssp, flags);
}
@@ -1382,7 +1436,9 @@ early_initcall(srcu_bootup_announce);
void __init srcu_init(void)
{
- struct srcu_struct *ssp;
+ struct srcu_struct *ssp, *tmp;
+
+ lockdep_assert_irqs_disabled();
/*
* Once that is set, call_srcu() can follow the normal path and
@@ -1390,12 +1446,13 @@ void __init srcu_init(void)
* and timers initialization.
*/
srcu_init_done = true;
- while (!list_empty(&srcu_boot_list)) {
- ssp = list_first_entry(&srcu_boot_list, struct srcu_struct,
- work.work.entry);
- list_del_init(&ssp->work.work.entry);
- queue_work(rcu_gp_wq, &ssp->work.work);
- }
+ /*
+ * ssp's that got initialized before rcu_init_geometry() have a stale
+ * node hierarchy. Rebuild their node trees and requeue their pending
+ * callbacks.
+ */
+ list_for_each_entry_safe(ssp, tmp, &srcu_early_init_list, early_init)
+ reset_srcu_struct(ssp);
}
#ifdef CONFIG_MODULES
--
2.25.1
