1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
|
From: Jens Axboe <axboe@kernel.dk>
Date: Wed, 4 Aug 2021 08:43:43 -0600
Subject: [PATCH] io-wq: remove GFP_ATOMIC allocation off schedule out path
Daniel reports that the v5.14-rc4-rt4 kernel throws a BUG when running
stress-ng:
| [ 90.202543] BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:35
| [ 90.202549] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 2047, name: iou-wrk-2041
| [ 90.202555] CPU: 5 PID: 2047 Comm: iou-wrk-2041 Tainted: G W 5.14.0-rc4-rt4+ #89
| [ 90.202559] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
| [ 90.202561] Call Trace:
| [ 90.202577] dump_stack_lvl+0x34/0x44
| [ 90.202584] ___might_sleep.cold+0x87/0x94
| [ 90.202588] rt_spin_lock+0x19/0x70
| [ 90.202593] ___slab_alloc+0xcb/0x7d0
| [ 90.202598] ? newidle_balance.constprop.0+0xf5/0x3b0
| [ 90.202603] ? dequeue_entity+0xc3/0x290
| [ 90.202605] ? io_wqe_dec_running.isra.0+0x98/0xe0
| [ 90.202610] ? pick_next_task_fair+0xb9/0x330
| [ 90.202612] ? __schedule+0x670/0x1410
| [ 90.202615] ? io_wqe_dec_running.isra.0+0x98/0xe0
| [ 90.202618] kmem_cache_alloc_trace+0x79/0x1f0
| [ 90.202621] io_wqe_dec_running.isra.0+0x98/0xe0
| [ 90.202625] io_wq_worker_sleeping+0x37/0x50
| [ 90.202628] schedule+0x30/0xd0
| [ 90.202630] schedule_timeout+0x8f/0x1a0
| [ 90.202634] ? __bpf_trace_tick_stop+0x10/0x10
| [ 90.202637] io_wqe_worker+0xfd/0x320
| [ 90.202641] ? finish_task_switch.isra.0+0xd3/0x290
| [ 90.202644] ? io_worker_handle_work+0x670/0x670
| [ 90.202646] ? io_worker_handle_work+0x670/0x670
| [ 90.202649] ret_from_fork+0x22/0x30
which is due to the RT kernel not liking a GFP_ATOMIC allocation inside
a raw spinlock. Besides that not working on RT, doing any kind of
allocation from inside schedule() is kind of nasty and should be avoided
if at all possible.
This particular path happens when an io-wq worker goes to sleep, and we
need a new worker to handle pending work. We currently allocate a small
data item to hold the information we need to create a new worker, but we
can instead include this data in the io_worker struct itself and just
protect it with a single bit lock. We only really need one per worker
anyway, as we will have run pending work between to sleep cycles.
https://lore.kernel.org/lkml/20210804082418.fbibprcwtzyt5qax@beryllium.lan/
Reported-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Tested-by: Daniel Wagner <dwagner@suse.de>
Link: https://lore.kernel.org/r/a673a130-e0e4-5aa8-4165-f35d1262fc6a@kernel.dk
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
fs/io-wq.c | 75 ++++++++++++++++++++++++++++++++++---------------------------
1 file changed, 42 insertions(+), 33 deletions(-)
--- a/fs/io-wq.c
+++ b/fs/io-wq.c
@@ -51,6 +51,10 @@ struct io_worker {
struct completion ref_done;
+ unsigned long create_state;
+ struct callback_head create_work;
+ int create_index;
+
struct rcu_head rcu;
};
@@ -270,50 +274,54 @@ static void io_wqe_inc_running(struct io
atomic_inc(&acct->nr_running);
}
-struct create_worker_data {
- struct callback_head work;
- struct io_wqe *wqe;
- int index;
-};
-
static void create_worker_cb(struct callback_head *cb)
{
- struct create_worker_data *cwd;
+ struct io_worker *worker;
struct io_wq *wq;
struct io_wqe *wqe;
struct io_wqe_acct *acct;
- cwd = container_of(cb, struct create_worker_data, work);
- wqe = cwd->wqe;
- wq = wqe->wq;
- acct = &wqe->acct[cwd->index];
+ worker = container_of(cb, struct io_worker, create_work);
+ wqe = worker->wqe;
+ wq = worker->wq;
+ acct = &wqe->acct[worker->create_index];
raw_spin_lock_irq(&wqe->lock);
if (acct->nr_workers < acct->max_workers)
acct->nr_workers++;
raw_spin_unlock_irq(&wqe->lock);
- create_io_worker(wq, cwd->wqe, cwd->index);
- kfree(cwd);
+ create_io_worker(wq, wqe, worker->create_index);
+ clear_bit_unlock(0, &worker->create_state);
+ io_worker_release(worker);
}
-static void io_queue_worker_create(struct io_wqe *wqe, struct io_wqe_acct *acct)
+static void io_queue_worker_create(struct io_wqe *wqe, struct io_worker
+ *worker, struct io_wqe_acct *acct)
{
- struct create_worker_data *cwd;
struct io_wq *wq = wqe->wq;
/* raced with exit, just ignore create call */
if (test_bit(IO_WQ_BIT_EXIT, &wq->state))
goto fail;
+ if (!io_worker_get(worker))
+ goto fail;
+ /*
+ * create_state manages ownership of create_work/index. We should
+ * only need one entry per worker, as the worker going to sleep
+ * will trigger the condition, and waking will clear it once it
+ * runs the task_work.
+ */
+ if (test_bit(0, &worker->create_state) ||
+ test_and_set_bit_lock(0, &worker->create_state))
+ goto fail_release;
+
+ init_task_work(&worker->create_work, create_worker_cb);
+ worker->create_index = acct->index;
+ if (!task_work_add(wq->task, &worker->create_work, TWA_SIGNAL))
+ return;
- cwd = kmalloc(sizeof(*cwd), GFP_ATOMIC);
- if (cwd) {
- init_task_work(&cwd->work, create_worker_cb);
- cwd->wqe = wqe;
- cwd->index = acct->index;
- if (!task_work_add(wq->task, &cwd->work, TWA_SIGNAL))
- return;
-
- kfree(cwd);
- }
+ clear_bit_unlock(0, &worker->create_state);
+fail_release:
+ io_worker_release(worker);
fail:
atomic_dec(&acct->nr_running);
io_worker_ref_put(wq);
@@ -331,7 +339,7 @@ static void io_wqe_dec_running(struct io
if (atomic_dec_and_test(&acct->nr_running) && io_wqe_run_queue(wqe)) {
atomic_inc(&acct->nr_running);
atomic_inc(&wqe->wq->worker_refs);
- io_queue_worker_create(wqe, acct);
+ io_queue_worker_create(wqe, worker, acct);
}
}
@@ -992,12 +1000,12 @@ struct io_wq *io_wq_create(unsigned boun
static bool io_task_work_match(struct callback_head *cb, void *data)
{
- struct create_worker_data *cwd;
+ struct io_worker *worker;
if (cb->func != create_worker_cb)
return false;
- cwd = container_of(cb, struct create_worker_data, work);
- return cwd->wqe->wq == data;
+ worker = container_of(cb, struct io_worker, create_work);
+ return worker->wqe->wq == data;
}
void io_wq_exit_start(struct io_wq *wq)
@@ -1014,12 +1022,13 @@ static void io_wq_exit_workers(struct io
return;
while ((cb = task_work_cancel_match(wq->task, io_task_work_match, wq)) != NULL) {
- struct create_worker_data *cwd;
+ struct io_worker *worker;
- cwd = container_of(cb, struct create_worker_data, work);
- atomic_dec(&cwd->wqe->acct[cwd->index].nr_running);
+ worker = container_of(cb, struct io_worker, create_work);
+ atomic_dec(&worker->wqe->acct[worker->create_index].nr_running);
io_worker_ref_put(wq);
- kfree(cwd);
+ clear_bit_unlock(0, &worker->create_state);
+ io_worker_release(worker);
}
rcu_read_lock();
|
