ipv4: fix a RCU-list lock in fib_triestat_seq_show
Subsystems (3)
PATCH v2 1/1 ipv4: fix a RCU-list lock in fib_triestat_seq_show
fib_triestat_seq_show() calls hlist_for_each_entry_rcu(tb, head, tb_hlist) without rcu_read_lock() will trigger a warning,
net/ipv4/fib_trie.c:2579 RCU-list traversed in non-reader section!!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 1
1 lock held by proc01/115277:
#0: c0000014507acf00 (&p->lock){+.+.}-{3:3}, at: seq_read+0x58/0x670
Call Trace: dump_stack+0xf4/0x164 (unreliable) lockdep_rcu_suspicious+0x140/0x164 fib_triestat_seq_show+0x750/0x880 seq_read+0x1a0/0x670 proc_reg_read+0x10c/0x1b0 __vfs_read+0x3c/0x70 vfs_read+0xac/0x170 ksys_read+0x7c/0x140 system_call+0x5c/0x68
Fix it by adding a pair of rcu_read_lock/unlock() and use cond_resched_rcu() to avoid the situation where walking of a large number of items may prevent scheduling for a long time.
---
Use cond_resched_rcu() from Eric.
net/ipv4/fib_trie.c | 3 +++ 1 file changed, 3 insertions(+)
net/ipv4/fib_trie.c +3 -0
@@ -2577,6 +2577,7 @@ static int fib_triestat_seq_show(struct seq_file *seq, void *v)" %zd bytes, size of tnode: %zd bytes.\n" , LEAF_SIZE, TNODE_SIZE(0)); +rcu_read_lock ();for (h =0 ; h <FIB_TABLE_HASHSZ ; h++) {struct hlist_head *head = &net->ipv4.fib_table_hash[h];struct fib_table *tb;
@@ -2596,7 +2597,9 @@ static int fib_triestat_seq_show(struct seq_file *seq, void *v)trie_show_usage (seq, t->stats);#endif } +cond_resched_rcu (); } +rcu_read_unlock ();return 0 ; }
Signature · 1 line
2.21.0 (Apple Git-122.2)
David Miller
maintainerFrom: Qian Cai <[email protected]>
Date: Wed, 25 Mar 2020 18:01:00 -0400
Quoted 22 lines
fib_triestat_seq_show()callshlist_for_each_entry_rcu(tb, head, tb_hlist)withoutrcu_read_lock()will trigger a warning,
net/ipv4/fib_trie.c:2579RCU-list traversed in non-reader section!!other info that might help us debug this:
rcu_scheduler_active= 2,debug_locks= 1
1 lock held by proc01/115277:
#0:c0000014507acf00(&p->lock){+.+.}-{3:3}, at:seq_read+0x58/0x670Call Trace: dump_stack+0xf4/0x164 (unreliable) lockdep_rcu_suspicious+0x140/0x164 fib_triestat_seq_show+0x750/0x880 seq_read+0x1a0/0x670 proc_reg_read+0x10c/0x1b0 __vfs_read+0x3c/0x70 vfs_read+0xac/0x170 ksys_read+0x7c/0x140 system_call+0x5c/0x68
Quoted 5 lines
Fix it by adding a pair of rcu_read_lock/unlock() and use
cond_resched_rcu()to avoid the situation where walking of a large number of items may prevent scheduling for a long time.Signed-off-by Qian Cai <[email protected]>
Eric, please review.
Eric Dumazet
Reviewed-byQuoted 11 lines · David Miller
From: Qian Cai <[email protected]>
Date: Wed, 25 Mar 2020 18:01:00 -0400Quoted 5 lines
Fix it by adding a pair of rcu_read_lock/unlock() and use
cond_resched_rcu()to avoid the situation where walking of a large number of items may prevent scheduling for a long time.Signed-off-by Qian Cai <[email protected]>Eric, please review.
Patch looks good to me, thanks !
David Miller
maintainerFrom: Qian Cai <[email protected]>
Date: Wed, 25 Mar 2020 18:01:00 -0400
Quoted 22 lines
fib_triestat_seq_show()callshlist_for_each_entry_rcu(tb, head, tb_hlist)withoutrcu_read_lock()will trigger a warning,
net/ipv4/fib_trie.c:2579RCU-list traversed in non-reader section!!other info that might help us debug this:
rcu_scheduler_active= 2,debug_locks= 1
1 lock held by proc01/115277:
#0:c0000014507acf00(&p->lock){+.+.}-{3:3}, at:seq_read+0x58/0x670Call Trace: dump_stack+0xf4/0x164 (unreliable) lockdep_rcu_suspicious+0x140/0x164 fib_triestat_seq_show+0x750/0x880 seq_read+0x1a0/0x670 proc_reg_read+0x10c/0x1b0 __vfs_read+0x3c/0x70 vfs_read+0xac/0x170 ksys_read+0x7c/0x140 system_call+0x5c/0x68
Quoted 5 lines
Fix it by adding a pair of rcu_read_lock/unlock() and use
cond_resched_rcu()to avoid the situation where walking of a large number of items may prevent scheduling for a long time.Signed-off-by Qian Cai <[email protected]>
Applied and queued up for -stable, thanks.