* [RFC 00/14] DEPT(DEPendency Tracker)
@ 2022-01-25 2:36 Byungchul Park
2022-01-25 2:36 ` [RFC 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (15 more replies)
0 siblings, 16 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Hi forks,
I've been developing a tool for detecting deadlock possibilities by
tracking wait/event rather than lock(?) acquisition order to try to
cover all synchonization machanisms. It's done on v5.10 tag. I bet
it would work great! Try it and see what's gonna happen.
Now that there's a porting issue, I made Dept rely on Lockdep. But it
should be separated from Lockdep once it's considered worth having.
Benifit:
0. Works with all lock primitives.
1. Works with wait_for_completion()/complete().
2. Works with 'wait' on PG_locked.
3. Works with 'wait' on PG_writeback.
4. Works with swait/wakeup.
5. Multiple reports are allowed.
6. Deduplication control on multiple reports.
7. Withstand false positives thanks to 5.
8. Easy to tag any wait/event.
Future work:
0. To make it more stable.
1. To separates Dept from Lockdep.
2. To improves performance in terms of time and space.
3. To use Dept as a dependency engine for Lockdep.
4. To add any missing tags of wait/event in the kernel.
5. To deduplicate stack trace.
I hope you guys are gonna be satisfied with Dept. Don't hesitate to
give any feedback. I will adopt any feedbacks if reasonable.
Thanks,
Byungchul
Byungchul Park (14):
llist: Move llist_{head,node} definition to types.h
dept: Implement Dept(Dependency Tracker)
dept: Embed Dept data in Lockdep
dept: Apply Dept to spinlock
dept: Apply Dept to mutex families
dept: Apply Dept to rwlock
dept: Apply Dept to wait_for_completion()/complete()
dept: Apply Dept to seqlock
dept: Apply Dept to rwsem
dept: Add proc knobs to show stats and dependency graph
dept: Introduce split map concept and new APIs for them
dept: Apply Dept to wait/event of PG_{locked,writeback}
dept: Separate out SDT(Single-event Dependency Tracker) header
dept: Apply SDT to swait
include/linux/completion.h | 48 +-
include/linux/dept.h | 541 ++++++++
include/linux/dept_page.h | 71 +
include/linux/dept_sdt.h | 53 +
include/linux/hardirq.h | 3 +
include/linux/irqflags.h | 33 +-
include/linux/llist.h | 9 +-
include/linux/lockdep.h | 156 ++-
include/linux/lockdep_types.h | 3 +
include/linux/mutex.h | 31 +
include/linux/page-flags.h | 26 +-
include/linux/pagemap.h | 7 +-
include/linux/percpu-rwsem.h | 10 +-
include/linux/rtmutex.h | 11 +-
include/linux/rwlock.h | 48 +
include/linux/rwlock_api_smp.h | 8 +-
include/linux/rwlock_types.h | 7 +
include/linux/rwsem.h | 31 +
include/linux/sched.h | 3 +
include/linux/seqlock.h | 19 +-
include/linux/spinlock.h | 24 +
include/linux/spinlock_types.h | 10 +
include/linux/swait.h | 4 +
include/linux/types.h | 8 +
init/init_task.c | 2 +
init/main.c | 4 +
kernel/Makefile | 1 +
kernel/dependency/Makefile | 5 +
kernel/dependency/dept.c | 2593 +++++++++++++++++++++++++++++++++++++
kernel/dependency/dept_hash.h | 11 +
kernel/dependency/dept_internal.h | 26 +
kernel/dependency/dept_object.h | 14 +
kernel/dependency/dept_proc.c | 97 ++
kernel/exit.c | 1 +
kernel/fork.c | 2 +
kernel/locking/lockdep.c | 12 +-
kernel/module.c | 2 +
kernel/sched/completion.c | 12 +-
kernel/sched/swait.c | 8 +
kernel/softirq.c | 6 +-
kernel/trace/trace_preemptirq.c | 19 +-
lib/Kconfig.debug | 21 +
mm/filemap.c | 62 +
mm/page_ext.c | 5 +
44 files changed, 4009 insertions(+), 58 deletions(-)
create mode 100644 include/linux/dept.h
create mode 100644 include/linux/dept_page.h
create mode 100644 include/linux/dept_sdt.h
create mode 100644 kernel/dependency/Makefile
create mode 100644 kernel/dependency/dept.c
create mode 100644 kernel/dependency/dept_hash.h
create mode 100644 kernel/dependency/dept_internal.h
create mode 100644 kernel/dependency/dept_object.h
create mode 100644 kernel/dependency/dept_proc.c
--
1.9.1
^ permalink raw reply [flat|nested] 31+ messages in thread
* [RFC 01/14] llist: Move llist_{head,node} definition to types.h
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 02/14] dept: Implement Dept(Dependency Tracker) Byungchul Park
` (14 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
llist_head and llist_node can be used by very primitives. For example,
Dept for tracking dependency uses llist things in its header. To avoid
header dependency, move those to types.h.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/llist.h | 9 +--------
include/linux/types.h | 8 ++++++++
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/include/linux/llist.h b/include/linux/llist.h
index 2e9c721..e0c88d5 100644
--- a/include/linux/llist.h
+++ b/include/linux/llist.h
@@ -50,14 +50,7 @@
#include <linux/atomic.h>
#include <linux/kernel.h>
-
-struct llist_head {
- struct llist_node *first;
-};
-
-struct llist_node {
- struct llist_node *next;
-};
+#include <linux/types.h>
#define LLIST_HEAD_INIT(name) { NULL }
#define LLIST_HEAD(name) struct llist_head name = LLIST_HEAD_INIT(name)
diff --git a/include/linux/types.h b/include/linux/types.h
index a147977..0b058f5 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -187,6 +187,14 @@ struct hlist_node {
struct hlist_node *next, **pprev;
};
+struct llist_head {
+ struct llist_node *first;
+};
+
+struct llist_node {
+ struct llist_node *next;
+};
+
struct ustat {
__kernel_daddr_t f_tfree;
__kernel_ino_t f_tinode;
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 02/14] dept: Implement Dept(Dependency Tracker)
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
2022-01-25 2:36 ` [RFC 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 03/14] dept: Embed Dept data in Lockdep Byungchul Park
` (13 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
CURRENT STATUS
--------------
Lockdep tracks acquisition order of locks in order to detect deadlock,
and IRQ and IRQ enable/disable state as well to take accident
acquisitions into account.
Lockdep should be turned off once it detects and reports a deadlock
since the data structure and algorithm are not reusable after detection
because of the complex design.
PROBLEM
-------
*Waits* and their *events* that never reach eventually cause deadlock.
However, Lockdep is only interested in lock acquisition order, forcing
to emulate lock acqusition even for just waits and events that have
nothing to do with real lock.
Even worse, no one likes Lockdep's false positive detection because that
prevents further one that might be more valuable. That's why all the
kernel developers are sensitive to Lockdep's false positive.
Besides those, by tracking acquisition order, it cannot correctly deal
with read lock and cross-event e.g. wait_for_completion()/complete() for
deadlock detection. Lockdep is no longer a good tool for that purpose.
SOLUTION
--------
Again, *waits* and their *events* that never reach eventually cause
deadlock. The new solution, Dept(DEPendency Tracker), focuses on waits
and events themselves. Dept tracks waits and events and report it if
any event would be never reachable.
Dept does:
. Works with read lock in the right way.
. Works with any wait and event e.i. cross-event.
. Continue to work even after reporting multiple times.
. Provides simple and intuitive APIs.
. Does exactly what dependency checker should do.
Q & A
-----
Q. Is this the first try ever to address the problem?
A. No. Cross-release feature (b09be676e0ff2 locking/lockdep: Implement
the 'crossrelease' feature) addressed it 2 years ago that was a
Lockdep extention and merged but reverted shortly because:
Cross-release started to report valuable hidden problems but started
to give report false positive reports as well. For sure, no one
likes Lockdep's false positive reports since it makes Lockdep stop,
preventing reporting further real problems.
Q. Why not Dept was developed as an extention of Lockdep?
A. Lockdep definitely includes all the efforts great developers have
made for a long time so as to be quite stable enough. But I had to
design and implement newly because of the following:
1) Lockdep was designed to track lock acquisition order. The APIs and
implementation do not fit on wait-event model.
2) Lockdep is turned off on detection including false positive. Which
is terrible and prevents developing any extention for stronger
detection.
Q. Do you intend to totally replace Lockdep?
A. No. Lockdep also checks if lock usage is correct. Of course, the
dependency check routine should be replaced but the other functions
should be still there.
Q. Do you mean the dependency check routine should be replaced right
away?
A. No. I admit Lockdep is stable enough thanks to great efforts kernel
developers have made. Lockdep and Dept, both should be in the kernel
until Dept gets considered stable.
Q. Stronger detection capability would give more false positive report.
Which was a big problem when cross-release was introduced. Is it ok
with Dept?
A. It's ok. Dept allows multiple reporting thanks to simple and quite
generalized design. Of course, false positive reports should be fixed
anyway but it's no longer as a critical problem as it was.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept.h | 541 +++++++++
include/linux/hardirq.h | 3 +
include/linux/irqflags.h | 33 +-
include/linux/sched.h | 3 +
init/init_task.c | 2 +
init/main.c | 2 +
kernel/Makefile | 1 +
kernel/dependency/Makefile | 4 +
kernel/dependency/dept.c | 2477 +++++++++++++++++++++++++++++++++++++++
kernel/dependency/dept_hash.h | 11 +
kernel/dependency/dept_object.h | 14 +
kernel/exit.c | 1 +
kernel/fork.c | 2 +
kernel/module.c | 2 +
kernel/softirq.c | 6 +-
kernel/trace/trace_preemptirq.c | 19 +-
lib/Kconfig.debug | 20 +
17 files changed, 3132 insertions(+), 9 deletions(-)
create mode 100644 include/linux/dept.h
create mode 100644 kernel/dependency/Makefile
create mode 100644 kernel/dependency/dept.c
create mode 100644 kernel/dependency/dept_hash.h
create mode 100644 kernel/dependency/dept_object.h
diff --git a/include/linux/dept.h b/include/linux/dept.h
new file mode 100644
index 0000000..8665629
--- /dev/null
+++ b/include/linux/dept.h
@@ -0,0 +1,541 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept(DEPendency Tracker) - runtime dependency tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ */
+
+#ifndef __LINUX_DEPT_H
+#define __LINUX_DEPT_H
+
+#ifdef CONFIG_DEPT
+
+#include <linux/types.h>
+
+struct task_struct;
+
+#define DEPT_MAX_STACK_ENTRY 16
+#define DEPT_MAX_WAIT_HIST 64
+#define DEPT_MAX_ECXT_HELD 48
+
+#define DEPT_MAX_SUBCLASSES 16
+#define DEPT_MAX_SUBCLASSES_EVT 2
+#define DEPT_MAX_SUBCLASSES_USR (DEPT_MAX_SUBCLASSES / DEPT_MAX_SUBCLASSES_EVT)
+#define DEPT_MAX_SUBCLASSES_CACHE 2
+
+#define DEPT_SIRQ 0
+#define DEPT_HIRQ 1
+#define DEPT_IRQS_NR 2
+#define DEPT_SIRQF (1UL << DEPT_SIRQ)
+#define DEPT_HIRQF (1UL << DEPT_HIRQ)
+
+struct dept_class {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * unique information about the class
+ */
+ const char *name;
+ unsigned long key;
+ int sub;
+
+ /*
+ * for BFS
+ */
+ unsigned int bfs_gen;
+ int bfs_dist;
+ struct dept_class *bfs_parent;
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking all classes
+ */
+ struct list_head all_node;
+
+ /*
+ * for associating its dependencies
+ */
+ struct list_head dep_head;
+ struct list_head dep_rev_head;
+
+ /*
+ * for tracking IRQ dependencies
+ */
+ int iwait_dist[DEPT_IRQS_NR];
+ struct dept_ecxt *iecxt[DEPT_IRQS_NR];
+ struct dept_wait *iwait[DEPT_IRQS_NR];
+};
+
+struct dept_stack {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * backtrace entries
+ */
+ unsigned long raw[DEPT_MAX_STACK_ENTRY];
+ int nr;
+};
+
+struct dept_ecxt {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * function that entered to this ecxt
+ */
+ const char *ecxt_fn;
+
+ /*
+ * event function
+ */
+ const char *event_fn;
+
+ /*
+ * associated class
+ */
+ struct dept_class *class;
+
+ /*
+ * flag indicating which IRQ has been
+ * enabled within the event context
+ */
+ unsigned long enirqf;
+
+ /*
+ * where the IRQ-enabled happened
+ */
+ unsigned long enirq_ip[DEPT_IRQS_NR];
+ struct dept_stack *enirq_stack[DEPT_IRQS_NR];
+
+ /*
+ * where the event context started
+ */
+ unsigned long ecxt_ip;
+ struct dept_stack *ecxt_stack;
+
+ /*
+ * where the event triggered
+ */
+ unsigned long event_ip;
+ struct dept_stack *event_stack;
+};
+
+struct dept_wait {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * function causing this wait
+ */
+ const char *wait_fn;
+
+ /*
+ * the associated class
+ */
+ struct dept_class *class;
+
+ /*
+ * which IRQ the wait was placed in
+ */
+ unsigned long irqf;
+
+ /*
+ * where the IRQ wait happened
+ */
+ unsigned long irq_ip[DEPT_IRQS_NR];
+ struct dept_stack *irq_stack[DEPT_IRQS_NR];
+
+ /*
+ * where the wait happened
+ */
+ unsigned long wait_ip;
+ struct dept_stack *wait_stack;
+};
+
+struct dept_staleiw {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking all staleiws
+ */
+ struct list_head all_node;
+
+ /*
+ * actual data
+ */
+ unsigned long ip;
+ int irq;
+};
+
+struct dept_staleie {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking all staleies
+ */
+ struct list_head all_node;
+
+ /*
+ * actual data
+ */
+ unsigned long ip;
+ int irq;
+};
+
+struct dept_dep {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * key data of dependency
+ */
+ struct dept_ecxt *ecxt;
+ struct dept_wait *wait;
+
+ /*
+ * This object can be referred without dept_lock
+ * held but with IRQ disabled, e.g. for hash
+ * lookup. So deferred deletion is needed.
+ */
+ struct rcu_head rh;
+
+ /*
+ * for BFS
+ */
+ struct list_head bfs_node;
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking to a class object
+ */
+ struct list_head dep_node;
+ struct list_head dep_rev_node;
+};
+
+struct dept_hash {
+ /*
+ * hash table
+ */
+ struct hlist_head *table;
+
+ /*
+ * size of the table e.i. 2^bits
+ */
+ int bits;
+};
+
+struct dept_pool {
+ const char *name;
+
+ /*
+ * object size
+ */
+ size_t obj_sz;
+
+ /*
+ * the number of the static array
+ */
+ atomic_t obj_nr;
+
+ /*
+ * offset of ->pool_node
+ */
+ size_t node_off;
+
+ /*
+ * pointer to the pool
+ */
+ void *spool;
+ struct llist_head boot_pool;
+ struct llist_head __percpu *lpool;
+};
+
+struct dept_ecxt_held {
+ /*
+ * associated event context
+ */
+ struct dept_ecxt *ecxt;
+
+ /*
+ * unique key for this dept_ecxt_held
+ */
+ unsigned long key;
+
+ /*
+ * the wgen when the event context started
+ */
+ unsigned int wgen;
+
+ /*
+ * for allowing user aware nesting
+ */
+ int nest;
+};
+
+struct dept_wait_hist {
+ /*
+ * associated wait
+ */
+ struct dept_wait *wait;
+
+ /*
+ * unique id of all waits system-wise until wrapped
+ */
+ unsigned int wgen;
+
+ /*
+ * local context id to identify IRQ context
+ */
+ unsigned int ctxt_id;
+};
+
+struct dept_task {
+ /*
+ * all event contexts that have entered and before exiting
+ */
+ struct dept_ecxt_held ecxt_held[DEPT_MAX_ECXT_HELD];
+ int ecxt_held_pos;
+
+ /*
+ * ring buffer holding all waits that have happened
+ */
+ struct dept_wait_hist wait_hist[DEPT_MAX_WAIT_HIST];
+ int wait_hist_pos;
+
+ /*
+ * sequencial id to identify each IRQ context
+ */
+ unsigned int irq_id[DEPT_IRQS_NR];
+
+ /*
+ * for tracking IRQ-enabled points with cross-event
+ */
+ unsigned int wgen_enirq[DEPT_IRQS_NR];
+
+ /*
+ * for keeping up-to-date IRQ-enabled points
+ */
+ unsigned long enirq_ip[DEPT_IRQS_NR];
+
+ /*
+ * current effective IRQ-enabled flag
+ */
+ unsigned long eff_enirqf;
+
+ /*
+ * for reserving a current stack instance at each operation
+ */
+ struct dept_stack *stack;
+
+ /*
+ * for preventing recursive call into Dept engine
+ */
+ int recursive;
+
+ /*
+ * for tracking IRQ-enable state
+ */
+ bool hardirqs_enabled;
+ bool softirqs_enabled;
+};
+
+struct dept_key {
+ union {
+ /*
+ * Each byte-wise address will be used as its key.
+ */
+ char subkeys[DEPT_MAX_SUBCLASSES];
+
+ /*
+ * for caching the main class pointer
+ */
+ struct dept_class *classes[DEPT_MAX_SUBCLASSES_CACHE];
+ };
+};
+
+struct dept_map {
+ const char *name;
+ struct dept_key *keys;
+ int sub_usr;
+
+ /*
+ * It's local copy for fast acces to the associated classes. And
+ * Also used for dept_key instance for statically defined map.
+ */
+ struct dept_key keys_local;
+
+ /*
+ * wait timestamp associated to this map
+ */
+ unsigned int wgen;
+
+ /*
+ * whether this map should be going to be checked or not
+ */
+ bool nocheck;
+};
+
+#define DEPT_TASK_INITIALIZER(t) \
+ .dept_task.wait_hist = { { .wait = NULL, } }, \
+ .dept_task.ecxt_held_pos = 0, \
+ .dept_task.wait_hist_pos = 0, \
+ .dept_task.irq_id = { 0 }, \
+ .dept_task.wgen_enirq = { 0 }, \
+ .dept_task.enirq_ip = { 0 }, \
+ .dept_task.recursive = 0, \
+ .dept_task.hardirqs_enabled = false, \
+ .dept_task.softirqs_enabled = false,
+
+extern void dept_on(void);
+extern void dept_off(void);
+extern void dept_init(void);
+extern void dept_task_init(struct task_struct *t);
+extern void dept_task_exit(struct task_struct *t);
+extern void dept_free_range(void *start, unsigned int sz);
+extern void dept_map_init(struct dept_map *m, struct dept_key *k, int sub, const char *n);
+extern void dept_map_reinit(struct dept_map *m);
+extern void dept_map_nocheck(struct dept_map *m);
+
+extern void dept_wait(struct dept_map *m, unsigned long w_f, unsigned long ip, const char *w_fn, int ne);
+extern void dept_ecxt_enter(struct dept_map *m, unsigned long e_f, unsigned long ip, const char *c_fn, const char *e_fn, int ne);
+extern void dept_asked_event(struct dept_map *m);
+extern void dept_event(struct dept_map *m, unsigned long e_f, unsigned long ip, const char *e_fn);
+extern void dept_ecxt_exit(struct dept_map *m, unsigned long ip);
+extern struct dept_map *dept_top_map(void);
+extern void dept_warn_on(bool cond);
+
+/*
+ * for users who want to manage external keys
+ */
+extern void dept_key_init(struct dept_key *k);
+extern void dept_key_destroy(struct dept_key *k);
+
+#define DEPT_SDT_MAP_INIT(dname) { .name = #dname }
+#define DEFINE_DEPT_SDT(x) \
+ struct dept_map x = DEPT_SDT_MAP_INIT(x)
+
+/*
+ * SDT(Single-event Dependency Tracker) APIs
+ *
+ * In case that one dept_map instance maps to a single event, SDT APIs
+ * can be used.
+ */
+#define sdt_map_init(m) \
+ do { \
+ static struct dept_key __key; \
+ dept_map_init(m, &__key, 0, #m); \
+ } while (0)
+#define sdt_map_init_key(m, k) dept_map_init(m, k, 0, #m)
+
+#define sdt_wait(m) \
+ do { \
+ dept_asked_event(m); \
+ dept_wait(m, 1UL, _THIS_IP_, "wait", 0); \
+ } while (0)
+#define sdt_ecxt_enter(m) dept_ecxt_enter(m, 1UL, _THIS_IP_, "start", "event", 0)
+#define sdt_event(m) dept_event(m, 1UL, _THIS_IP_, "event")
+#define sdt_ecxt_exit(m) dept_ecxt_exit(m, _THIS_IP_)
+#else /* !CONFIG_DEPT */
+struct dept_task { };
+struct dept_key { };
+struct dept_map { };
+
+#define DEPT_TASK_INITIALIZER(t)
+
+#define dept_on() do { } while (0)
+#define dept_off() do { } while (0)
+#define dept_init() do { } while (0)
+#define dept_task_init(t) do { } while (0)
+#define dept_task_exit(t) do { } while (0)
+#define dept_free_range(s, sz) do { } while (0)
+#define dept_map_init(m, k, s, n) do { (void)(n); (void)(k); } while (0)
+#define dept_map_reinit(m) do { } while (0)
+#define dept_map_nocheck(m) do { } while (0)
+
+#define dept_wait(m, w_f, ip, w_fn, ne) do { (void)(w_fn); } while (0)
+#define dept_ecxt_enter(m, e_f, ip, c_fn, e_fn, ne) do { (void)(c_fn); (void)(e_fn); } while (0)
+#define dept_asked_event(m) do { } while (0)
+#define dept_event(m, e_f, ip, e_fn) do { (void)(e_fn); } while (0)
+#define dept_ecxt_exit(m, ip) do { } while (0)
+#define dept_top_map() NULL
+#define dept_warn_on(c) do { } while (0)
+#define dept_key_init(k) do { (void)(k); } while (0)
+#define dept_key_destroy(k) do { (void)(k); } while (0)
+
+#define DEPT_SDT_MAP_INIT(dname)
+#define DEFINE_DEPT_SDT(x)
+
+#define sdt_map_init(m) do { } while (0)
+#define sdt_map_init_key(m, k) do { (void)(k); } while (0)
+#define sdt_wait(m) do { } while (0)
+#define sdt_ecxt_enter(m) do { } while (0)
+#define sdt_event(m) do { } while (0)
+#define sdt_ecxt_exit(m) do { } while (0)
+#endif
+#endif /* __LINUX_DEPT_H */
diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index 754f67a..55798fb 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -5,6 +5,7 @@
#include <linux/context_tracking_state.h>
#include <linux/preempt.h>
#include <linux/lockdep.h>
+#include <linux/dept.h>
#include <linux/ftrace_irq.h>
#include <linux/vtime.h>
#include <asm/hardirq.h>
@@ -113,6 +114,7 @@ static inline void rcu_nmi_exit(void) { }
*/
#define __nmi_enter() \
do { \
+ dept_off(); \
lockdep_off(); \
arch_nmi_enter(); \
printk_nmi_enter(); \
@@ -137,6 +139,7 @@ static inline void rcu_nmi_exit(void) { }
printk_nmi_exit(); \
arch_nmi_exit(); \
lockdep_on(); \
+ dept_on(); \
} while (0)
#define nmi_exit() \
diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index 3ed4e87..dbab683 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -31,6 +31,22 @@
static inline void lockdep_hardirqs_off(unsigned long ip) { }
#endif
+#ifdef CONFIG_DEPT
+ extern void dept_hardirq_enter(void);
+ extern void dept_softirq_enter(void);
+ extern void dept_enable_hardirq(unsigned long ip);
+ extern void dept_enable_softirq(unsigned long ip);
+ extern void dept_disable_hardirq(unsigned long ip);
+ extern void dept_disable_softirq(unsigned long ip);
+#else
+ static inline void dept_hardirq_enter(void) { }
+ static inline void dept_softirq_enter(void) { }
+ static inline void dept_enable_hardirq(unsigned long ip) { }
+ static inline void dept_enable_softirq(unsigned long ip) { }
+ static inline void dept_disable_hardirq(unsigned long ip) { }
+ static inline void dept_disable_softirq(unsigned long ip) { }
+#endif
+
#ifdef CONFIG_TRACE_IRQFLAGS
/* Per-task IRQ trace events information. */
@@ -53,15 +69,19 @@ struct irqtrace_events {
extern void trace_hardirqs_off_finish(void);
extern void trace_hardirqs_on(void);
extern void trace_hardirqs_off(void);
+extern void trace_softirqs_on_caller(unsigned long ip);
+extern void trace_softirqs_off_caller(unsigned long ip);
# define lockdep_hardirq_context() (raw_cpu_read(hardirq_context))
# define lockdep_softirq_context(p) ((p)->softirq_context)
# define lockdep_hardirqs_enabled() (this_cpu_read(hardirqs_enabled))
# define lockdep_softirqs_enabled(p) ((p)->softirqs_enabled)
-# define lockdep_hardirq_enter() \
-do { \
- if (__this_cpu_inc_return(hardirq_context) == 1)\
- current->hardirq_threaded = 0; \
+# define lockdep_hardirq_enter() \
+do { \
+ if (__this_cpu_inc_return(hardirq_context) == 1) { \
+ current->hardirq_threaded = 0; \
+ dept_hardirq_enter(); \
+ } \
} while (0)
# define lockdep_hardirq_threaded() \
do { \
@@ -73,7 +93,8 @@ struct irqtrace_events {
} while (0)
# define lockdep_softirq_enter() \
do { \
- current->softirq_context++; \
+ if (!current->softirq_context++) \
+ dept_softirq_enter(); \
} while (0)
# define lockdep_softirq_exit() \
do { \
@@ -123,6 +144,8 @@ struct irqtrace_events {
# define trace_hardirqs_off_finish() do { } while (0)
# define trace_hardirqs_on() do { } while (0)
# define trace_hardirqs_off() do { } while (0)
+# define trace_softirqs_on_caller(ip) do { } while (0)
+# define trace_softirqs_off_caller(ip) do { } while (0)
# define lockdep_hardirq_context() 0
# define lockdep_softirq_context(p) 0
# define lockdep_hardirqs_enabled() 0
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 76cd21f..c056933 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -34,6 +34,7 @@
#include <linux/rseq.h>
#include <linux/seqlock.h>
#include <linux/kcsan.h>
+#include <linux/dept.h>
/* task_struct member predeclarations (sorted alphabetically): */
struct audit_context;
@@ -1035,6 +1036,8 @@ struct task_struct {
struct held_lock held_locks[MAX_LOCK_DEPTH];
#endif
+ struct dept_task dept_task;
+
#if defined(CONFIG_UBSAN) && !defined(CONFIG_UBSAN_TRAP)
unsigned int in_ubsan;
#endif
diff --git a/init/init_task.c b/init/init_task.c
index a56f0ab..3bd94e6 100644
--- a/init/init_task.c
+++ b/init/init_task.c
@@ -12,6 +12,7 @@
#include <linux/audit.h>
#include <linux/numa.h>
#include <linux/scs.h>
+#include <linux/dept.h>
#include <linux/uaccess.h>
@@ -197,6 +198,7 @@ struct task_struct init_task
.curr_chain_key = INITIAL_CHAIN_KEY,
.lockdep_recursion = 0,
#endif
+ DEPT_TASK_INITIALIZER(init_task)
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
.ret_stack = NULL,
#endif
diff --git a/init/main.c b/init/main.c
index 32b2a8a..1818a90 100644
--- a/init/main.c
+++ b/init/main.c
@@ -63,6 +63,7 @@
#include <linux/debug_locks.h>
#include <linux/debugobjects.h>
#include <linux/lockdep.h>
+#include <linux/dept.h>
#include <linux/kmemleak.h>
#include <linux/padata.h>
#include <linux/pid_namespace.h>
@@ -991,6 +992,7 @@ asmlinkage __visible void __init __no_sanitize_address start_kernel(void)
panic_param);
lockdep_init();
+ dept_init();
/*
* Need to run this when irqs are enabled, because it wants
diff --git a/kernel/Makefile b/kernel/Makefile
index 6c9f199..9fb8388 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -47,6 +47,7 @@ obj-y += rcu/
obj-y += livepatch/
obj-y += dma/
obj-y += entry/
+obj-y += dependency/
obj-$(CONFIG_CHECKPOINT_RESTORE) += kcmp.o
obj-$(CONFIG_FREEZER) += freezer.o
diff --git a/kernel/dependency/Makefile b/kernel/dependency/Makefile
new file mode 100644
index 0000000..9f7778e
--- /dev/null
+++ b/kernel/dependency/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_DEPT) += dept.o
+
diff --git a/kernel/dependency/dept.c b/kernel/dependency/dept.c
new file mode 100644
index 0000000..64287894
--- /dev/null
+++ b/kernel/dependency/dept.c
@@ -0,0 +1,2477 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept(DEPendency Tracker) - Runtime dependency tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ *
+ * Dept provides a general way to detect deadlock possibility in runtime
+ * and the interest is not limited to typical lock but to every
+ * syncronization primitives.
+ *
+ * The following ideas were borrowed from Lockdep:
+ *
+ * 1) Use a graph to track relationship between classes.
+ * 2) Prevent performance regression using hash.
+ *
+ * The following items were enhanced from Lockdep:
+ *
+ * 1) Cover more deadlock cases.
+ * 2) Allow muliple reports.
+ *
+ * TODO: Both Lockdep and Dept should co-exist until Dept is considered
+ * stable. Then the dependency check routine shoud be replaced with Dept
+ * after. It should finally look like:
+ *
+ *
+ *
+ * As is:
+ *
+ * Lockdep
+ * +-----------------------------------------+
+ * | Lock usage correctness check | <-> locks
+ * | |
+ * | |
+ * | +-------------------------------------+ |
+ * | | Dependency check | |
+ * | | (by tracking lock aquisition order) | |
+ * | +-------------------------------------+ |
+ * | |
+ * +-----------------------------------------+
+ *
+ * Dept
+ * +-----------------------------------------+
+ * | Dependency check | <-> waits/events
+ * | (by tracking wait and event context) |
+ * +-----------------------------------------+
+ *
+ *
+ *
+ * To be:
+ *
+ * Lockdep
+ * +-----------------------------------------+
+ * | Lock usage correctness check | <-> locks
+ * | |
+ * | |
+ * | (Request dependency check) |
+ * | T |
+ * +--------------------|--------------------+
+ * |
+ * Dept V
+ * +-----------------------------------------+
+ * | Dependency check | <-> waits/events
+ * | (by tracking wait and event context) |
+ * +-----------------------------------------+
+ *
+ *
+ *
+ * ---
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your ootion) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ */
+
+#include <linux/sched.h>
+#include <linux/stacktrace.h>
+#include <linux/spinlock.h>
+#include <linux/kallsyms.h>
+#include <linux/hash.h>
+#include <linux/dept.h>
+#include <linux/utsname.h>
+
+static int dept_stop;
+static int dept_per_cpu_ready;
+
+/*
+ * Make all operations using DEPT_WARN_ON() fail on oops_in_progress and
+ * prevent warning message.
+ */
+#define DEPT_WARN_ON_ONCE(c) \
+ ({ \
+ int __ret = 1; \
+ if (likely(!oops_in_progress)) \
+ __ret = WARN_ONCE(c, "DEPT_WARN_ON_ONCE: " #c); \
+ __ret; \
+ })
+
+#define DEPT_WARN_ONCE(s...) \
+ ({ \
+ if (likely(!oops_in_progress)) \
+ WARN_ONCE(1, "DEPT_WARN_ONCE: " s); \
+ })
+
+#define DEPT_WARN_ON(c) \
+ ({ \
+ int __ret = 1; \
+ if (likely(!oops_in_progress)) \
+ __ret = WARN(c, "DEPT_WARN_ON: " #c); \
+ __ret; \
+ })
+
+#define DEPT_WARN(s...) \
+ ({ \
+ if (likely(!oops_in_progress)) \
+ WARN(1, "DEPT_WARN: " s); \
+ })
+
+#define DEPT_STOP(s...) \
+ ({ \
+ WRITE_ONCE(dept_stop, 1); \
+ if (likely(!oops_in_progress)) \
+ WARN(1, "DEPT_STOP: " s); \
+ })
+
+static arch_spinlock_t dept_spin = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;
+
+/*
+ * Dept internal engine should be careful in using outside functions
+ * e.g. printk at reporting since that kind of usage might cause
+ * untrackable deadlock.
+ */
+static atomic_t dept_outworld = ATOMIC_INIT(0);
+
+static inline void dept_outworld_enter(void)
+{
+ atomic_inc(&dept_outworld);
+}
+
+static inline void dept_outworld_exit(void)
+{
+ atomic_dec(&dept_outworld);
+}
+
+static inline bool dept_outworld_entered(void)
+{
+ return atomic_read(&dept_outworld);
+}
+
+static inline bool dept_lock(void)
+{
+ while (!arch_spin_trylock(&dept_spin))
+ if (unlikely(dept_outworld_entered()))
+ return false;
+ return true;
+}
+
+static inline void dept_unlock(void)
+{
+ arch_spin_unlock(&dept_spin);
+}
+
+/*
+ * whether to stack-trace on every wait or every ecxt
+ */
+static bool rich_stack = true;
+
+enum bfs_ret {
+ DEPT_BFS_CONTINUE,
+ DEPT_BFS_CONTINUE_REV,
+ DEPT_BFS_DONE,
+ DEPT_BFS_SKIP,
+};
+
+/*
+ * The irq wait is in unknown state. Should identify the state.
+ */
+#define DEPT_IWAIT_UNKNOWN ((void *)1)
+
+static inline bool before(unsigned int a, unsigned int b)
+{
+ return (int)(a - b) < 0;
+}
+
+static inline bool valid_stack(struct dept_stack *s)
+{
+ return s && s->nr > 0;
+}
+
+static inline bool valid_class(struct dept_class *c)
+{
+ return c->key;
+}
+
+static inline void invalidate_class(struct dept_class *c)
+{
+ c->key = 0UL;
+}
+
+static inline struct dept_class *dep_fc(struct dept_dep *d)
+{
+ return d->ecxt->class;
+}
+
+static inline struct dept_class *dep_tc(struct dept_dep *d)
+{
+ return d->wait->class;
+}
+
+static inline const char *irq_str(int irq)
+{
+ if (irq == DEPT_SIRQ)
+ return "softirq";
+ if (irq == DEPT_HIRQ)
+ return "hardirq";
+ return "(unknown)";
+}
+
+static inline struct dept_task *dept_task(void)
+{
+ return ¤t->dept_task;
+}
+
+/*
+ * Pool
+ * =====================================================================
+ * Dept maintains pools to provide objects in a safe way.
+ *
+ * 1) Static pool is used at the beginning of booting time.
+ * 2) Local pool is tried first before the static pool. Objects that
+ * have been freed will be placed.
+ */
+
+enum object_t {
+#define OBJECT(id, nr) OBJECT_##id,
+ #include "dept_object.h"
+#undef OBJECT
+ OBJECT_NR,
+};
+
+#define OBJECT(id, nr) \
+static struct dept_##id spool_##id[nr]; \
+static DEFINE_PER_CPU(struct llist_head, lpool_##id);
+ #include "dept_object.h"
+#undef OBJECT
+
+static struct dept_pool pool[OBJECT_NR] = {
+#define OBJECT(id, nr) { \
+ .name = #id, \
+ .obj_sz = sizeof(struct dept_##id), \
+ .obj_nr = ATOMIC_INIT(nr), \
+ .node_off = offsetof(struct dept_##id, pool_node), \
+ .spool = spool_##id, \
+ .lpool = &lpool_##id, },
+ #include "dept_object.h"
+#undef OBJECT
+};
+
+/*
+ * Can use llist no matter whether CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG is
+ * enabled because Dept never race with NMI by nesting control.
+ */
+static void *from_pool(enum object_t t)
+{
+ struct dept_pool *p;
+ struct llist_head *h;
+ struct llist_node *n;
+
+ /*
+ * llist_del_first() doesn't allow concurrent access e.g.
+ * between process and IRQ context.
+ */
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return NULL;
+
+ p = &pool[t];
+
+ /*
+ * Try local pool first.
+ */
+ if (likely(dept_per_cpu_ready))
+ h = this_cpu_ptr(p->lpool);
+ else
+ h = &p->boot_pool;
+
+ n = llist_del_first(h);
+ if (n)
+ return (void *)n - p->node_off;
+
+ /*
+ * Try static pool.
+ */
+ if (atomic_read(&p->obj_nr) > 0) {
+ int idx = atomic_dec_return(&p->obj_nr);
+ if (idx >= 0)
+ return p->spool + (idx * p->obj_sz);
+ }
+
+ DEPT_WARN_ONCE("Pool(%s) is empty.\n", p->name);
+ return NULL;
+}
+
+static void to_pool(void *o, enum object_t t)
+{
+ struct dept_pool *p = &pool[t];
+ struct llist_head *h;
+
+ preempt_disable();
+ if (likely(dept_per_cpu_ready))
+ h = this_cpu_ptr(p->lpool);
+ else
+ h = &p->boot_pool;
+
+ llist_add(o + p->node_off, h);
+ preempt_enable();
+}
+
+#define OBJECT(id, nr) \
+static void (*ctor_##id)(struct dept_##id *a); \
+static void (*dtor_##id)(struct dept_##id *a); \
+static inline struct dept_##id *new_##id(void) \
+{ \
+ struct dept_##id *a; \
+ \
+ a = (struct dept_##id *)from_pool(OBJECT_##id); \
+ if (unlikely(!a)) \
+ return NULL; \
+ \
+ atomic_set(&a->ref, 1); \
+ \
+ if (ctor_##id) \
+ ctor_##id(a); \
+ \
+ return a; \
+} \
+ \
+static inline struct dept_##id *get_##id(struct dept_##id *a) \
+{ \
+ atomic_inc(&a->ref); \
+ return a; \
+} \
+ \
+static inline void put_##id(struct dept_##id *a) \
+{ \
+ if (!atomic_dec_return(&a->ref)) { \
+ if (dtor_##id) \
+ dtor_##id(a); \
+ to_pool(a, OBJECT_##id); \
+ } \
+} \
+ \
+static inline void del_##id(struct dept_##id *a) \
+{ \
+ put_##id(a); \
+} \
+ \
+static inline bool id##_refered(struct dept_##id *a, int expect) \
+{ \
+ return a && atomic_read(&a->ref) > expect; \
+}
+#include "dept_object.h"
+#undef OBJECT
+
+#define SET_CONSTRUCTOR(id, f) \
+static void (*ctor_##id)(struct dept_##id *a) = f;
+
+static void initialize_dep(struct dept_dep *d)
+{
+ INIT_LIST_HEAD(&d->bfs_node);
+ INIT_LIST_HEAD(&d->dep_node);
+ INIT_LIST_HEAD(&d->dep_rev_node);
+}
+SET_CONSTRUCTOR(dep, initialize_dep);
+
+static void initialize_class(struct dept_class *c)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ c->iwait_dist[i] = INT_MAX;
+ WRITE_ONCE(c->iecxt[i], NULL);
+ WRITE_ONCE(c->iwait[i], NULL);
+ }
+ c->bfs_gen = 0U;
+
+ INIT_LIST_HEAD(&c->all_node);
+ INIT_LIST_HEAD(&c->dep_head);
+ INIT_LIST_HEAD(&c->dep_rev_head);
+}
+SET_CONSTRUCTOR(class, initialize_class);
+
+static void initialize_ecxt(struct dept_ecxt *e)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ e->enirq_stack[i] = NULL;
+ e->enirq_ip[i] = 0UL;
+ }
+ e->ecxt_ip = 0UL;
+ e->ecxt_stack = NULL;
+ e->enirqf = 0UL;
+ e->event_stack = NULL;
+}
+SET_CONSTRUCTOR(ecxt, initialize_ecxt);
+
+static void initialize_wait(struct dept_wait *w)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ w->irq_stack[i] = NULL;
+ w->irq_ip[i] = 0UL;
+ }
+ w->wait_ip = 0UL;
+ w->wait_stack = NULL;
+ w->irqf = 0UL;
+}
+SET_CONSTRUCTOR(wait, initialize_wait);
+
+static void initialize_stack(struct dept_stack *s)
+{
+ s->nr = 0;
+}
+SET_CONSTRUCTOR(stack, initialize_stack);
+
+#define OBJECT(id, nr) \
+static void (*ctor_##id)(struct dept_##id *a);
+ #include "dept_object.h"
+#undef OBJECT
+
+#undef SET_CONSTRUCTOR
+
+#define SET_DESTRUCTOR(id, f) \
+static void (*dtor_##id)(struct dept_##id *a) = f;
+
+static void destroy_dep(struct dept_dep *d)
+{
+ if (d->ecxt)
+ put_ecxt(d->ecxt);
+ if (d->wait)
+ put_wait(d->wait);
+}
+SET_DESTRUCTOR(dep, destroy_dep);
+
+static void destroy_ecxt(struct dept_ecxt *e)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++)
+ if (e->enirq_stack[i])
+ put_stack(e->enirq_stack[i]);
+ if (e->class)
+ put_class(e->class);
+ if (e->ecxt_stack)
+ put_stack(e->ecxt_stack);
+ if (e->event_stack)
+ put_stack(e->event_stack);
+}
+SET_DESTRUCTOR(ecxt, destroy_ecxt);
+
+static void destroy_wait(struct dept_wait *w)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++)
+ if (w->irq_stack[i])
+ put_stack(w->irq_stack[i]);
+ if (w->class)
+ put_class(w->class);
+ if (w->wait_stack)
+ put_stack(w->wait_stack);
+}
+SET_DESTRUCTOR(wait, destroy_wait);
+
+#define OBJECT(id, nr) \
+static void (*dtor_##id)(struct dept_##id *a);
+ #include "dept_object.h"
+#undef OBJECT
+
+#undef SET_DESTRUCTOR
+
+/*
+ * Caching and hashing
+ * =====================================================================
+ * Dept makes use of caching and hashing to improve performance. Each
+ * object can be obtained in O(1) with its key.
+ *
+ * NOTE: Currently we assume all the objects in the hashs will never be
+ * removed. Implement it when needed.
+ */
+
+/*
+ * Some information might be lost but it's only for hashing key.
+ */
+static inline unsigned long mix(unsigned long a, unsigned long b)
+{
+ int halfbits = sizeof(unsigned long) * 8 / 2;
+ unsigned long halfmask = (1UL << halfbits) - 1UL;
+ return (a << halfbits) | (b & halfmask);
+}
+
+static bool cmp_dep(struct dept_dep *d1, struct dept_dep *d2)
+{
+ return dep_fc(d1)->key == dep_fc(d2)->key &&
+ dep_tc(d1)->key == dep_tc(d2)->key;
+}
+
+static unsigned long key_dep(struct dept_dep *d)
+{
+ return mix(dep_fc(d)->key, dep_tc(d)->key);
+}
+
+static bool cmp_class(struct dept_class *c1, struct dept_class *c2)
+{
+ return c1->key == c2->key;
+}
+
+static unsigned long key_class(struct dept_class *c)
+{
+ return c->key;
+}
+
+static bool cmp_staleiw(struct dept_staleiw *s1, struct dept_staleiw *s2)
+{
+ return s1->ip == s2->ip && s1->irq == s2->irq;
+}
+
+static unsigned long key_staleiw(struct dept_staleiw *s)
+{
+ return s->ip ^ (1UL << s->irq);
+}
+
+static bool cmp_staleie(struct dept_staleie *s1, struct dept_staleie *s2)
+{
+ return s1->ip == s2->ip && s1->irq == s2->irq;
+}
+
+static unsigned long key_staleie(struct dept_staleie *s)
+{
+ return s->ip ^ (1UL << s->irq);
+}
+
+#define HASH(id, bits) \
+static struct hlist_head table_##id[1UL << bits]; \
+ \
+static inline struct hlist_head *head_##id(struct dept_##id *a) \
+{ \
+ return table_##id + hash_long(key_##id(a), bits); \
+} \
+ \
+static inline struct dept_##id *hash_lookup_##id(struct dept_##id *a) \
+{ \
+ struct dept_##id *b; \
+ hlist_for_each_entry_rcu(b, head_##id(a), hash_node) \
+ if (cmp_##id(a, b)) \
+ return b; \
+ return NULL; \
+} \
+ \
+static inline void hash_add_##id(struct dept_##id *a) \
+{ \
+ hlist_add_head_rcu(&a->hash_node, head_##id(a)); \
+} \
+ \
+static inline void hash_del_##id(struct dept_##id *a) \
+{ \
+ hlist_del_rcu(&a->hash_node); \
+}
+#include "dept_hash.h"
+#undef HASH
+
+static inline struct dept_dep *dep(struct dept_class *fc,
+ struct dept_class *tc)
+{
+ struct dept_ecxt onetime_e = { .class = fc };
+ struct dept_wait onetime_w = { .class = tc };
+ struct dept_dep onetime_d = { .ecxt = &onetime_e,
+ .wait = &onetime_w };
+ return hash_lookup_dep(&onetime_d);
+}
+
+static inline struct dept_class *class(unsigned long key)
+{
+ struct dept_class onetime_c = { .key = key };
+ return hash_lookup_class(&onetime_c);
+}
+
+static inline struct dept_staleiw *staleiw(unsigned long ip, int irq)
+{
+ struct dept_staleiw onetime_s = { .ip = ip, .irq = irq };
+ return hash_lookup_staleiw(&onetime_s);
+}
+
+static inline struct dept_staleie *staleie(unsigned long ip, int irq)
+{
+ struct dept_staleie onetime_s = { .ip = ip, .irq = irq };
+ return hash_lookup_staleie(&onetime_s);
+}
+
+/*
+ * Report
+ * =====================================================================
+ * Dept prints useful information to help debuging on detection of
+ * problematic dependency.
+ */
+
+static inline void print_ip_stack(unsigned long ip, struct dept_stack *s)
+{
+ if (ip)
+ print_ip_sym(KERN_WARNING, ip);
+
+ if (valid_stack(s)) {
+ printk("stacktrace:\n");
+ stack_trace_print(s->raw, s->nr, 5);
+ }
+
+ if (!ip && !valid_stack(s))
+ printk("(N/A)\n");
+}
+
+#define print_spc(spc, fmt, ...) \
+ printk("%*c" fmt, (spc) * 4, ' ', ##__VA_ARGS__)
+
+static void print_diagram(struct dept_dep *d)
+{
+ struct dept_ecxt *e = d->ecxt;
+ struct dept_wait *w = d->wait;
+ struct dept_class *fc = dep_fc(d);
+ struct dept_class *tc = dep_tc(d);
+ unsigned long irqf;
+ int irq;
+ bool firstline = true;
+ int spc = 1;
+ const char *w_fn = w->wait_fn ?: "(unknown)";
+ const char *e_fn = e->event_fn ?: "(unknown)";
+ const char *c_fn = e->ecxt_fn ?: "(unknown)";
+
+ irqf = e->enirqf & w->irqf;
+
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR) {
+ if (!firstline)
+ printk("\nor\n\n");
+ firstline = false;
+
+ print_spc(spc, "[S] %s(%s:%d)\n", c_fn, fc->name, fc->sub);
+ print_spc(spc, " <%s interrupt>\n", irq_str(irq));
+ print_spc(spc + 1, "[W] %s(%s:%d)\n", w_fn, tc->name, tc->sub);
+ print_spc(spc, "[E] %s(%s:%d)\n", e_fn, fc->name, fc->sub);
+ }
+
+ if (!irqf) {
+ print_spc(spc, "[S] %s(%s:%d)\n", c_fn, fc->name, fc->sub);
+ print_spc(spc, "[W] %s(%s:%d)\n", w_fn, tc->name, tc->sub);
+ print_spc(spc, "[E] %s(%s:%d)\n", e_fn, fc->name, fc->sub);
+ }
+}
+
+static void print_dep(struct dept_dep *d)
+{
+ struct dept_ecxt *e = d->ecxt;
+ struct dept_wait *w = d->wait;
+ struct dept_class *fc = dep_fc(d);
+ struct dept_class *tc = dep_tc(d);
+ unsigned long irqf;
+ int irq;
+ const char *w_fn = w->wait_fn ?: "(unknown)";
+ const char *e_fn = e->event_fn ?: "(unknown)";
+ const char *c_fn = e->ecxt_fn ?: "(unknown)";
+
+ irqf = e->enirqf & w->irqf;
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR) {
+ printk("%s has been enabled:\n", irq_str(irq));
+ print_ip_stack(e->enirq_ip[irq], e->enirq_stack[irq]);
+ printk("\n");
+
+ printk("[S] %s(%s:%d):\n", c_fn, fc->name, fc->sub);
+ print_ip_stack(e->ecxt_ip, e->ecxt_stack);
+ printk("\n");
+
+ printk("[W] %s(%s:%d) in %s context:\n",
+ w_fn, tc->name, tc->sub, irq_str(irq));
+ print_ip_stack(w->irq_ip[irq], w->irq_stack[irq]);
+ printk("\n");
+
+ printk("[E] %s(%s:%d):\n", e_fn, fc->name, fc->sub);
+ print_ip_stack(e->event_ip, e->event_stack);
+ }
+
+ if (!irqf) {
+ printk("[S] %s(%s:%d):\n", c_fn, fc->name, fc->sub);
+ print_ip_stack(e->ecxt_ip, e->ecxt_stack);
+ printk("\n");
+
+ printk("[W] %s(%s:%d):\n", w_fn, tc->name, tc->sub);
+ print_ip_stack(w->wait_ip, w->wait_stack);
+ printk("\n");
+
+ printk("[E] %s(%s:%d):\n", e_fn, fc->name, fc->sub);
+ print_ip_stack(e->event_ip, e->event_stack);
+ }
+}
+
+static void save_current_stack(int skip);
+
+/*
+ * Print all classes in a circle.
+ */
+static void print_circle(struct dept_class *c)
+{
+ struct dept_class *fc = c->bfs_parent;
+ struct dept_class *tc = c;
+ int i;
+
+ dept_outworld_enter();
+ save_current_stack(6);
+
+ printk("===================================================\n");
+ printk("Dept: Circular dependency has been detected.\n");
+ printk("%s %.*s %s\n", init_utsname()->release,
+ (int)strcspn(init_utsname()->version, " "),
+ init_utsname()->version,
+ print_tainted());
+ printk("---------------------------------------------------\n");
+ printk("summary\n");
+ printk("---------------------------------------------------\n");
+
+ if (fc == tc)
+ printk("*** AA DEADLOCK ***\n\n");
+ else
+ printk("*** DEADLOCK ***\n\n");
+
+ i = 0;
+ do {
+ struct dept_dep *d = dep(fc, tc);
+
+ printk("context %c\n", 'A' + (i++));
+ print_diagram(d);
+ if (fc != c)
+ printk("\n");
+
+ tc = fc;
+ fc = fc->bfs_parent;
+ } while (tc != c);
+
+ printk("\n");
+ printk("[S]: start of the event context\n");
+ printk("[W]: the wait blocked\n");
+ printk("[E]: the event not reachable\n");
+
+ i = 0;
+ do {
+ struct dept_dep *d = dep(fc, tc);
+
+ printk("---------------------------------------------------\n");
+ printk("context %c's detail\n", 'A' + i);
+ printk("---------------------------------------------------\n");
+ printk("context %c\n", 'A' + (i++));
+ print_diagram(d);
+ printk("\n");
+ print_dep(d);
+
+ tc = fc;
+ fc = fc->bfs_parent;
+ } while (tc != c);
+
+ printk("---------------------------------------------------\n");
+ printk("information that might be helpful\n");
+ printk("---------------------------------------------------\n");
+ dump_stack();
+
+ dept_outworld_exit();
+}
+
+/*
+ * BFS(Breadth First Search)
+ * =====================================================================
+ * Whenever a new dependency is added into the graph, search the graph
+ * for a new circular dependency.
+ */
+
+static inline void enqueue(struct list_head *h, struct dept_dep *d)
+{
+ list_add_tail(&d->bfs_node, h);
+}
+
+static inline struct dept_dep *dequeue(struct list_head *h)
+{
+ struct dept_dep *d;
+ d = list_first_entry(h, struct dept_dep, bfs_node);
+ list_del(&d->bfs_node);
+ return d;
+}
+
+static inline bool empty(struct list_head *h)
+{
+ return list_empty(h);
+}
+
+static void extend_queue(struct list_head *h, struct dept_class *cur)
+{
+ struct dept_dep *d;
+
+ list_for_each_entry(d, &cur->dep_head, dep_node) {
+ struct dept_class *next = dep_tc(d);
+ if (cur->bfs_gen == next->bfs_gen)
+ continue;
+ next->bfs_gen = cur->bfs_gen;
+ next->bfs_dist = cur->bfs_dist + 1;
+ next->bfs_parent = cur;
+ enqueue(h, d);
+ }
+}
+
+static void extend_queue_rev(struct list_head *h, struct dept_class *cur)
+{
+ struct dept_dep *d;
+
+ list_for_each_entry(d, &cur->dep_rev_head, dep_rev_node) {
+ struct dept_class *next = dep_fc(d);
+ if (cur->bfs_gen == next->bfs_gen)
+ continue;
+ next->bfs_gen = cur->bfs_gen;
+ next->bfs_dist = cur->bfs_dist + 1;
+ next->bfs_parent = cur;
+ enqueue(h, d);
+ }
+}
+
+typedef enum bfs_ret bfs_f(struct dept_dep *d, void *in, void **out);
+static unsigned int bfs_gen;
+
+/*
+ * NOTE: Must be called with dept_lock held.
+ */
+static void bfs(struct dept_class *c, bfs_f *cb, void *in, void **out)
+{
+ LIST_HEAD(q);
+ enum bfs_ret ret;
+
+ if (DEPT_WARN_ON(!cb))
+ return;
+
+ /*
+ * Avoid zero bfs_gen.
+ */
+ bfs_gen = bfs_gen + 1 ?: 1;
+
+ c->bfs_gen = bfs_gen;
+ c->bfs_dist = 0;
+ c->bfs_parent = c;
+
+ ret = cb(NULL, in, out);
+ if (ret == DEPT_BFS_DONE)
+ return;
+ if (ret == DEPT_BFS_SKIP)
+ return;
+ if (ret == DEPT_BFS_CONTINUE)
+ extend_queue(&q, c);
+ if (ret == DEPT_BFS_CONTINUE_REV)
+ extend_queue_rev(&q, c);
+
+ while (!empty(&q)) {
+ struct dept_dep *d = dequeue(&q);
+
+ ret = cb(d, in, out);
+ if (ret == DEPT_BFS_DONE)
+ break;
+ if (ret == DEPT_BFS_SKIP)
+ continue;
+ if (ret == DEPT_BFS_CONTINUE)
+ extend_queue(&q, dep_tc(d));
+ if (ret == DEPT_BFS_CONTINUE_REV)
+ extend_queue_rev(&q, dep_fc(d));
+ }
+
+ while (!empty(&q))
+ dequeue(&q);
+}
+
+/*
+ * Main operations
+ * =====================================================================
+ * Add dependencies - Each new dependency is added into the graph and
+ * checked if it forms a circular dependency.
+ *
+ * Track waits - Waits are queued into the ring buffer for later use to
+ * generate appropriate dependencies with cross-event.
+ *
+ * Track event contexts(ecxt) - Event contexts are pushed into local
+ * stack for later use to generate appropriate dependencies with waits.
+ */
+
+static inline unsigned long cur_enirqf(void);
+static inline int cur_irq(void);
+static inline unsigned int cur_ctxt_id(void);
+
+static inline struct dept_stack *get_current_stack(void)
+{
+ struct dept_stack *s = dept_task()->stack;
+ return s ? get_stack(s) : NULL;
+}
+
+static inline void prepare_current_stack(void)
+{
+ struct dept_stack *s = dept_task()->stack;
+ /*
+ * The reference counter would be 1 when no one refers to the
+ * dept_stack.
+ */
+ const int expect_ref = 1;
+
+ /*
+ * The dept_stack is already ready.
+ */
+ if (s && !stack_refered(s, expect_ref)) {
+ s->nr = 0;
+ return;
+ }
+
+ if (s)
+ put_stack(s);
+
+ s = dept_task()->stack = new_stack();
+ if (!s)
+ return;
+
+ get_stack(s);
+ del_stack(s);
+}
+
+static void save_current_stack(int skip)
+{
+ struct dept_stack *s = dept_task()->stack;
+
+ if (!s)
+ return;
+ if (valid_stack(s))
+ return;
+
+ s->nr = stack_trace_save(s->raw, DEPT_MAX_STACK_ENTRY, skip);
+}
+
+static void finish_current_stack(void)
+{
+ struct dept_stack *s = dept_task()->stack;
+ /*
+ * The reference counter would be 1 when no one refers to the
+ * dept_stack.
+ */
+ const int expect_ref = 1;
+
+ if (s && stack_refered(s, expect_ref))
+ save_current_stack(2);
+}
+
+/*
+ * FIXME: For now, disable Lockdep while Dept is working.
+ *
+ * Both Lockdep and Dept report it on a deadlock detection using
+ * printk taking the risk of another deadlock that might be caused by
+ * locks of console or printk between inside and outside of them.
+ *
+ * For Dept, it's no problem since multiple reports are allowed. But it
+ * would be a bad idea for Lockdep since it will stop even on a singe
+ * report. So we need to prevent Lockdep from its reporting the risk
+ * Dept would take when reporting something.
+ */
+#include <linux/lockdep.h>
+
+void dept_off(void)
+{
+ dept_task()->recursive++;
+ lockdep_off();
+}
+
+void dept_on(void)
+{
+ dept_task()->recursive--;
+ lockdep_on();
+}
+
+static inline unsigned long dept_enter(void)
+{
+ unsigned long flags;
+ raw_local_irq_save(flags);
+ dept_off();
+ prepare_current_stack();
+ return flags;
+}
+
+static inline void dept_exit(unsigned long flags)
+{
+ finish_current_stack();
+ dept_on();
+ raw_local_irq_restore(flags);
+}
+
+/*
+ * NOTE: Must be called with dept_lock held.
+ */
+static struct dept_dep *__add_dep(struct dept_ecxt *e,
+ struct dept_wait *w)
+{
+ struct dept_dep *d;
+
+ if (!valid_class(e->class) || !valid_class(w->class))
+ return NULL;
+
+ if (dep(e->class, w->class))
+ return NULL;
+
+ d = new_dep();
+ if (unlikely(!d))
+ return NULL;
+
+ d->ecxt = get_ecxt(e);
+ d->wait = get_wait(w);
+
+ /*
+ * Add the dependency into hash and graph.
+ */
+ hash_add_dep(d);
+ list_add(&d->dep_node , &dep_fc(d)->dep_head );
+ list_add(&d->dep_rev_node, &dep_tc(d)->dep_rev_head);
+ return d;
+}
+
+static enum bfs_ret cb_check_dl(struct dept_dep *d,
+ void *in, void **out)
+{
+ struct dept_dep *new = (struct dept_dep *)in;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d) {
+ dep_tc(new)->bfs_parent = dep_fc(new);
+
+ if (dep_tc(new) != dep_fc(new))
+ return DEPT_BFS_CONTINUE;
+
+ /*
+ * AA circle does not make additional deadlock. We don't
+ * have to continue this BFS search.
+ */
+ print_circle(dep_tc(new));
+ return DEPT_BFS_DONE;
+ }
+
+ /*
+ * Allow multiple reports.
+ */
+ if (dep_tc(d) == dep_fc(new))
+ print_circle(dep_tc(new));
+
+ return DEPT_BFS_CONTINUE;
+}
+
+/*
+ * This function is actually in charge of reporting.
+ */
+static inline void check_dl(struct dept_dep *d)
+{
+ bfs(dep_tc(d), cb_check_dl, (void *)d, NULL);
+}
+
+/*
+ * Should keep the search until the end even if it encounters
+ * iwait_dist <= bfs_dist on the way because there might be
+ * DEPT_IWAIT_UNKNOWN or something here and there which should be filled
+ * with a valid one.
+ *
+ * XXX: The search by the branch can stop when it encounters
+ * iwait_dist <= bfs_dist, if it's proved that there are no need to go
+ * in the case.
+ */
+static enum bfs_ret cb_prop_iwait(struct dept_dep *d,
+ void *in, void **out)
+{
+ int irq = *(int *)in;
+ struct dept_class *fc;
+ struct dept_class *tc;
+ struct dept_dep *new;
+
+ if (DEPT_WARN_ON(!out))
+ return DEPT_BFS_DONE;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d)
+ return DEPT_BFS_CONTINUE;
+
+ fc = dep_fc(d);
+ tc = dep_tc(d);
+
+ if (tc->iwait_dist[irq] <= tc->bfs_dist)
+ return DEPT_BFS_CONTINUE;
+
+ tc->iwait_dist[irq] = tc->bfs_dist;
+ WRITE_ONCE(tc->iwait[irq], fc->iwait[irq]);
+
+ if (!tc->iecxt[irq])
+ return DEPT_BFS_CONTINUE;
+
+ /*
+ * Add irq dependency to form a complete circle.
+ */
+ new = __add_dep(tc->iecxt[irq], tc->iwait[irq]);
+ if (!new)
+ return DEPT_BFS_CONTINUE;
+
+ *out = new;
+ return DEPT_BFS_DONE;
+}
+
+/*
+ * Should keep the search until the end even if it encounters another
+ * iwait on the way because there might be the iwait here and there by
+ * any chance which should be cleaned as well.
+ *
+ * XXX: The search by the branch can stop when it encounters another
+ * iwait, if it's proved that there are no more the iwait in the case.
+ */
+static enum bfs_ret cb_clean_iwait(struct dept_dep *d, void *in, void **out)
+{
+ struct dept_class *c = (struct dept_class *)in;
+ struct dept_class *tc;
+ int i;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d)
+ return DEPT_BFS_CONTINUE;
+
+ tc = dep_tc(d);
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ if (!tc->iwait[i] || tc->iwait[i] != c->iwait[i])
+ continue;
+ WRITE_ONCE(tc->iwait[i], DEPT_IWAIT_UNKNOWN);
+ tc->iwait_dist[i] = INT_MAX;
+ }
+
+ return DEPT_BFS_CONTINUE;
+}
+
+/*
+ * Should perform cleaning iwait even if c is not a root of the iwait
+ * because c might be a bridge for the iwait to propagate.
+ */
+static void clean_iwait(struct dept_class *root, struct dept_class *c)
+{
+ bfs(root, cb_clean_iwait, (void *)c, NULL);
+}
+
+static LIST_HEAD(staleies);
+static LIST_HEAD(staleiws);
+
+static void stale_iecxt_iwait(struct dept_dep *d, int irq)
+{
+ struct dept_ecxt *ie = d->ecxt;
+ struct dept_wait *iw = d->wait;
+ struct dept_class *iwroot = iw->class;
+ struct dept_staleie *sie;
+ struct dept_staleiw *siw;
+ struct dept_class onetime_c;
+ int i;
+
+ /*
+ * iwroot should be root of the iwait.
+ */
+ DEPT_WARN_ON(iwroot->iwait_dist[irq] != 0);
+
+ /*
+ * ie and iw should have one common irq flag at least.
+ */
+ DEPT_WARN_ON(!(ie->enirqf & iw->irqf));
+
+ for (i = 0; i < DEPT_IRQS_NR; i++)
+ onetime_c.iwait[i] = (i == irq) ? iw : NULL;
+ clean_iwait(iwroot, &onetime_c);
+
+ sie = new_staleie();
+ if (likely(sie)) {
+ sie->ip = ie->enirq_ip[irq];
+ sie->irq = irq;
+ hash_add_staleie(sie);
+ list_add(&sie->all_node, &staleies);
+ }
+ ie->enirqf &= ~(1UL << irq);
+ put_ecxt(ie);
+ WRITE_ONCE(dep_fc(d)->iecxt[irq], NULL);
+
+ siw = new_staleiw();
+ if (likely(siw)) {
+ siw->ip = iw->irq_ip[irq];
+ siw->irq = irq;
+ hash_add_staleiw(siw);
+ list_add(&siw->all_node, &staleiws);
+ }
+ iw->irqf &= ~(1UL << irq);
+ put_wait(iw);
+ WRITE_ONCE(dep_tc(d)->iwait[irq], NULL);
+}
+
+/*
+ * Should start from a class that has 0 of iwait_dist.
+ */
+static void propagate_iwait(struct dept_class *c, int irq)
+{
+ struct dept_dep *new;
+ do {
+ new = NULL;
+ bfs(c, cb_prop_iwait, (void *)&irq, (void **)&new);
+
+ /*
+ * Deadlock detected. Let check_dl() report it.
+ */
+ if (new) {
+ check_dl(new);
+ stale_iecxt_iwait(new, irq);
+ }
+ } while (new);
+}
+
+static enum bfs_ret cb_find_iwait(struct dept_dep *d, void *in, void **out)
+{
+ int irq = *(int *)in;
+ struct dept_wait *w;
+
+ if (DEPT_WARN_ON(!out))
+ return DEPT_BFS_DONE;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d)
+ return DEPT_BFS_CONTINUE_REV;
+
+ w = dep_fc(d)->iwait[irq];
+ if (!w || w == DEPT_IWAIT_UNKNOWN)
+ return DEPT_BFS_CONTINUE_REV;
+
+ *out = w;
+ return DEPT_BFS_DONE;
+}
+
+static struct dept_wait *find_iwait(struct dept_class *c, int irq)
+{
+ struct dept_wait *w = NULL;
+ bfs(c, cb_find_iwait, (void *)&irq, (void **)&w);
+ return w;
+}
+
+static void add_iecxt(struct dept_ecxt *e, int irq, bool stack)
+{
+ /*
+ * This access is safe since we ensure e->class has set locally.
+ */
+ struct dept_class *c = e->class;
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(c->iecxt[irq]))
+ return;
+
+ if (unlikely(staleie(dt->enirq_ip[irq], irq)))
+ return;
+
+ if (unlikely(!dept_lock()))
+ return;
+
+ if (c->iecxt[irq])
+ goto unlock;
+
+ if (unlikely(staleie(dt->enirq_ip[irq], irq)))
+ goto unlock;
+
+ e->enirqf |= (1UL << irq);
+
+ WRITE_ONCE(c->iecxt[irq], get_ecxt(e));
+
+ /*
+ * Should be NULL since it's the first time that these
+ * enirq_{ip,stack}[irq] have ever set.
+ */
+ DEPT_WARN_ON(e->enirq_ip[irq]);
+ DEPT_WARN_ON(e->enirq_stack[irq]);
+
+ e->enirq_ip[irq] = dt->enirq_ip[irq];
+ if (stack)
+ e->enirq_stack[irq] = get_current_stack();
+
+ if (c->iwait[irq] == DEPT_IWAIT_UNKNOWN) {
+ struct dept_wait *w = find_iwait(c, irq);
+
+ /*
+ * Deadlock detected. Let propagate_iwait() report it.
+ */
+ if (w)
+ propagate_iwait(w->class, irq);
+ } else if (c->iwait[irq]) {
+ /*
+ * Add irq dependency to form a complete circle.
+ */
+ struct dept_dep *d = __add_dep(e, c->iwait[irq]);
+
+ /*
+ * Deadlock detected. Let check_dl() report it.
+ */
+ if (d) {
+ check_dl(d);
+ stale_iecxt_iwait(d, irq);
+ }
+ }
+unlock:
+ dept_unlock();
+}
+
+static void add_iwait(struct dept_wait *w, int irq)
+{
+ struct dept_class *c = w->class;
+ struct dept_wait *iw;
+
+ iw = READ_ONCE(c->iwait[irq]);
+ if (iw && READ_ONCE(iw->class) == c)
+ return;
+
+ if (unlikely(staleiw(w->irq_ip[irq], irq)))
+ return;
+
+ if (unlikely(!dept_lock()))
+ return;
+
+ iw = c->iwait[irq];
+ if (iw && iw->class == c)
+ goto unlock;
+
+ if (unlikely(staleiw(w->irq_ip[irq], irq)))
+ goto unlock;
+
+ w->irqf |= (1UL << irq);
+
+ c->iwait_dist[irq] = 0;
+ WRITE_ONCE(c->iwait[irq], get_wait(w));
+
+ /*
+ * Should be NULL since it's the first time that these
+ * irq_{ip,stack}[irq] have ever set.
+ */
+ DEPT_WARN_ON(w->irq_ip[irq]);
+ DEPT_WARN_ON(w->irq_stack[irq]);
+
+ w->irq_ip[irq] = w->wait_ip;
+ w->irq_stack[irq] = get_current_stack();
+
+ if (c->iecxt[irq]) {
+ /*
+ * Add irq dependency to form a complete circle.
+ */
+ struct dept_dep *d = __add_dep(c->iecxt[irq], w);
+
+ /*
+ * Deadlock detected. Let check_dl() report it.
+ */
+ if (d) {
+ check_dl(d);
+ stale_iecxt_iwait(d, irq);
+ }
+ }
+ propagate_iwait(c, irq);
+unlock:
+ dept_unlock();
+}
+
+static inline struct dept_wait_hist *hist(int pos)
+{
+ struct dept_task *dt = dept_task();
+ return dt->wait_hist + (pos % DEPT_MAX_WAIT_HIST);
+}
+
+static inline int hist_pos_next(void)
+{
+ struct dept_task *dt = dept_task();
+ return dt->wait_hist_pos % DEPT_MAX_WAIT_HIST;
+}
+
+static inline void hist_advance(void)
+{
+ struct dept_task *dt = dept_task();
+ dt->wait_hist_pos++;
+ dt->wait_hist_pos %= DEPT_MAX_WAIT_HIST;
+}
+
+static inline struct dept_wait_hist *new_hist(void)
+{
+ struct dept_wait_hist *wh = hist(hist_pos_next());
+ hist_advance();
+ return wh;
+}
+
+static void add_hist(struct dept_wait *w, unsigned int wg, unsigned int ctxt_id)
+{
+ struct dept_wait_hist *wh = new_hist();
+
+ if (likely(wh->wait))
+ put_wait(wh->wait);
+
+ wh->wait = get_wait(w);
+ wh->wgen = wg;
+ wh->ctxt_id = ctxt_id;
+}
+
+static void add_dep(struct dept_ecxt *e, struct dept_wait *w)
+{
+ struct dept_dep *d;
+ int i;
+
+ if (dep(e->class, w->class))
+ return;
+
+ if (unlikely(!dept_lock()))
+ return;
+
+ d = __add_dep(e, w);
+ if (d) {
+ check_dl(d);
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ /*
+ * There's no iwait to propagate in the 'from'.
+ */
+ if (!e->class->iwait[i])
+ continue;
+ /*
+ * This case has been already handled by add_iwait().
+ */
+ if (w->class->iwait[i])
+ continue;
+
+ propagate_iwait(e->class->iwait[i]->class, i);
+ }
+ }
+ dept_unlock();
+}
+
+static atomic_t wgen = ATOMIC_INIT(1);
+
+static void add_wait(struct dept_class *c, unsigned long ip,
+ const char *w_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_wait *w;
+ unsigned int wg = 0U;
+ int irq;
+ int i;
+
+ w = new_wait();
+ if (unlikely(!w))
+ return;
+
+ WRITE_ONCE(w->class, get_class(c));
+ w->wait_ip = ip;
+ w->wait_fn = w_fn;
+ w->wait_stack = get_current_stack();
+
+ irq = cur_irq();
+ if (irq < DEPT_IRQS_NR)
+ add_iwait(w, irq);
+
+ /*
+ * Avoid adding dependency between user aware nested ecxt and
+ * wait.
+ */
+ for (i = dt->ecxt_held_pos - 1; i >= 0; i--) {
+ struct dept_ecxt_held *eh;
+ eh = dt->ecxt_held + i;
+ if (eh->ecxt->class != c || eh->nest == ne)
+ break;
+ }
+
+ for (; i >= 0; i--) {
+ struct dept_ecxt_held *eh;
+ eh = dt->ecxt_held + i;
+ add_dep(eh->ecxt, w);
+ }
+
+ if (!wait_refered(w, 1) && !rich_stack) {
+ if (w->wait_stack)
+ put_stack(w->wait_stack);
+ w->wait_stack = NULL;
+ }
+
+ /*
+ * Avoid zero wgen.
+ */
+ wg = atomic_inc_return(&wgen) ?: atomic_inc_return(&wgen);
+ add_hist(w, wg, cur_ctxt_id());
+
+ del_wait(w);
+}
+
+static void add_ecxt(void *obj, struct dept_class *c, unsigned long ip,
+ const char *c_fn, const char *e_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_ecxt_held *eh;
+ struct dept_ecxt *e;
+ unsigned long irqf;
+ int irq;
+
+ if (DEPT_WARN_ON(dt->ecxt_held_pos == DEPT_MAX_ECXT_HELD))
+ return;
+
+ e = new_ecxt();
+ if (unlikely(!e))
+ return;
+
+ e->class = get_class(c);
+ e->ecxt_ip = ip;
+ e->ecxt_stack = ip && rich_stack ? get_current_stack() : NULL;
+ e->event_fn = e_fn;
+ e->ecxt_fn = c_fn;
+
+ eh = dt->ecxt_held + (dt->ecxt_held_pos++);
+ eh->ecxt = get_ecxt(e);
+ eh->key = (unsigned long)obj;
+ eh->wgen = atomic_read(&wgen);
+ eh->nest = ne;
+
+ irqf = cur_enirqf();
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR)
+ add_iecxt(e, irq, false);
+
+ del_ecxt(e);
+}
+
+static int find_ecxt_pos(unsigned long key, bool newfirst)
+{
+ struct dept_task *dt = dept_task();
+ int i;
+
+ if (newfirst) {
+ for (i = dt->ecxt_held_pos - 1; i >= 0; i--)
+ if (dt->ecxt_held[i].key == key)
+ return i;
+ } else {
+ for (i = 0; i < dt->ecxt_held_pos; i++)
+ if (dt->ecxt_held[i].key == key)
+ return i;
+ }
+ return -1;
+}
+
+static void pop_ecxt(void *obj)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long key = (unsigned long)obj;
+ int pos;
+ int i;
+
+ /*
+ * TODO: WARN on pos == -1.
+ */
+ pos = find_ecxt_pos(key, true);
+ if (pos == -1)
+ return;
+
+ put_ecxt(dt->ecxt_held[pos].ecxt);
+ dt->ecxt_held_pos--;
+
+ for (i = pos; i < dt->ecxt_held_pos; i++)
+ dt->ecxt_held[i] = dt->ecxt_held[i + 1];
+}
+
+static inline bool good_hist(struct dept_wait_hist *wh, unsigned int wg)
+{
+ return wh->wait != NULL && before(wg, wh->wgen);
+}
+
+/*
+ * Binary-search the ring buffer for the earliest valid wait.
+ */
+static int find_hist_pos(unsigned int wg)
+{
+ int oldest;
+ int l;
+ int r;
+ int pos;
+
+ oldest = hist_pos_next();
+ if (unlikely(good_hist(hist(oldest), wg))) {
+ DEPT_WARN_ONCE("Need to expand the ring buffer.\n");
+ return oldest;
+ }
+
+ l = oldest + 1;
+ r = oldest + DEPT_MAX_WAIT_HIST - 1;
+ for (pos = (l + r) / 2; l <= r; pos = (l + r) / 2) {
+ struct dept_wait_hist *p = hist(pos - 1);
+ struct dept_wait_hist *wh = hist(pos);
+
+ if (!good_hist(p, wg) && good_hist(wh, wg))
+ return pos % DEPT_MAX_WAIT_HIST;
+ if (good_hist(wh, wg))
+ r = pos - 1;
+ else
+ l = pos + 1;
+ }
+ return -1;
+}
+
+static void do_event(void *obj, struct dept_class *c, unsigned int wg,
+ unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_wait_hist *wh;
+ struct dept_ecxt_held *eh;
+ unsigned long key = (unsigned long)obj;
+ unsigned int ctxt_id;
+ int end;
+ int pos;
+ int i;
+
+ /*
+ * The event was triggered before wait.
+ */
+ if (!wg)
+ return;
+
+ pos = find_ecxt_pos(key, false);
+ if (pos == -1)
+ return;
+
+ eh = dt->ecxt_held + pos;
+ eh->ecxt->event_ip = ip;
+ eh->ecxt->event_stack = get_current_stack();
+
+ /*
+ * The ecxt already has done what it needs.
+ */
+ if (!before(wg, eh->wgen))
+ return;
+
+ pos = find_hist_pos(wg);
+ if (pos == -1)
+ return;
+
+ ctxt_id = cur_ctxt_id();
+ end = hist_pos_next();
+ end = end > pos ? end : end + DEPT_MAX_WAIT_HIST;
+ for (wh = hist(pos); pos < end; wh = hist(++pos)) {
+ if (wh->ctxt_id == ctxt_id)
+ add_dep(eh->ecxt, wh->wait);
+ if (!before(wh->wgen, eh->wgen))
+ break;
+ }
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ if (before(dt->wgen_enirq[i], wg))
+ continue;
+ add_iecxt(eh->ecxt, i, false);
+ }
+}
+
+static void del_dep_rcu(struct rcu_head *rh)
+{
+ struct dept_dep *d = container_of(rh, struct dept_dep, rh);
+
+ preempt_disable();
+ del_dep(d);
+ preempt_enable();
+}
+
+/*
+ * NOTE: Must be called with dept_lock held.
+ */
+static void disconnect_class(struct dept_class *c)
+{
+ struct dept_dep *d, *n;
+ int i;
+
+ clean_iwait(c, c);
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ struct dept_ecxt *e = c->iecxt[i];
+ struct dept_wait *w = c->iwait[i];
+ if (e)
+ put_ecxt(e);
+ if (w && w->class == c)
+ put_wait(w);
+ WRITE_ONCE(c->iecxt[i], NULL);
+ WRITE_ONCE(c->iwait[i], NULL);
+ }
+
+ list_for_each_entry_safe(d, n, &c->dep_head, dep_node) {
+ list_del_rcu(&d->dep_node);
+ list_del_rcu(&d->dep_rev_node);
+ hash_del_dep(d);
+ call_rcu(&d->rh, del_dep_rcu);
+ }
+
+ list_for_each_entry_safe(d, n, &c->dep_rev_head, dep_rev_node) {
+ list_del_rcu(&d->dep_node);
+ list_del_rcu(&d->dep_rev_node);
+ hash_del_dep(d);
+ call_rcu(&d->rh, del_dep_rcu);
+ }
+}
+
+/*
+ * IRQ context control
+ * =====================================================================
+ * Whether a wait is in {hard,soft}-IRQ context or whether
+ * {hard,soft}-IRQ has been enabled on the way to an event is very
+ * important to check dependency. All those things should be tracked.
+ */
+
+static inline unsigned long cur_enirqf(void)
+{
+ struct dept_task *dt = dept_task();
+ int he = dt->hardirqs_enabled;
+ int se = dt->softirqs_enabled;
+ if (he)
+ return DEPT_HIRQF | (se ? DEPT_SIRQF : 0UL);
+ return 0UL;
+}
+
+static inline int cur_irq(void)
+{
+ if (lockdep_softirq_context(current))
+ return DEPT_SIRQ;
+ if (lockdep_hardirq_context())
+ return DEPT_HIRQ;
+ return DEPT_IRQS_NR;
+}
+
+static inline unsigned int cur_ctxt_id(void)
+{
+ struct dept_task *dt = dept_task();
+ int irq = cur_irq();
+
+ /*
+ * Normal process context
+ */
+ if (irq == DEPT_IRQS_NR)
+ return 0U;
+
+ return dt->irq_id[irq] | (1UL << irq);
+}
+
+static void enirq_transition(int irq)
+{
+ struct dept_task *dt = dept_task();
+ int i;
+
+ /*
+ * READ wgen >= wgen of an event with IRQ enabled has been
+ * observed on the way to the event means, the IRQ can cut in
+ * within the ecxt. Used for cross-event detection.
+ *
+ * wait context event context(ecxt)
+ * ------------ -------------------
+ * wait event
+ * WRITE wgen
+ * observe IRQ enabled
+ * READ wgen
+ * keep the wgen locally
+ *
+ * on the event
+ * check the local wgen
+ */
+ dt->wgen_enirq[irq] = atomic_read(&wgen);
+
+ for (i = dt->ecxt_held_pos - 1; i >= 0; i--) {
+ struct dept_ecxt_held *eh;
+ eh = dt->ecxt_held + i;
+ add_iecxt(eh->ecxt, irq, true);
+ }
+}
+
+static void enirq_update(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long irqf;
+ unsigned long prev;
+ int irq;
+
+ prev = dt->eff_enirqf;
+ irqf = cur_enirqf();
+ dt->eff_enirqf = irqf;
+
+ /*
+ * Do enirq_transition() only on an OFF -> ON transition.
+ */
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR) {
+ unsigned long flags;
+
+ if (prev & (1UL << irq))
+ continue;
+
+ flags = dept_enter();
+ dt->enirq_ip[irq] = ip;
+ enirq_transition(irq);
+ dept_exit(flags);
+ }
+}
+
+/*
+ * Ensure it has been called on OFF -> ON transition.
+ */
+void dept_enable_softirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(early_boot_irqs_disabled))
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->softirqs_enabled = true;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it has been called on OFF -> ON transition.
+ */
+void dept_enable_hardirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(early_boot_irqs_disabled))
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->hardirqs_enabled = true;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it has been called on ON -> OFF transition.
+ */
+void dept_disable_softirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->softirqs_enabled = false;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it has been called on ON -> OFF transition.
+ */
+void dept_disable_hardirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->hardirqs_enabled = false;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it's the outmost softirq context.
+ */
+void dept_softirq_enter(void)
+{
+ struct dept_task *dt = dept_task();
+ dt->irq_id[DEPT_SIRQ] += (1UL << DEPT_IRQS_NR);
+}
+
+/*
+ * Ensure it's the outmost hardirq context.
+ */
+void dept_hardirq_enter(void)
+{
+ struct dept_task *dt = dept_task();
+ dt->irq_id[DEPT_HIRQ] += (1UL << DEPT_IRQS_NR);
+}
+
+/*
+ * Dept API
+ * =====================================================================
+ * Main Dept APIs.
+ */
+
+static inline void clean_classes_cache(struct dept_key *k)
+{
+ int i;
+
+ for (i = 0; i < DEPT_MAX_SUBCLASSES_CACHE; i++)
+ k->classes[i] = NULL;
+}
+
+void dept_map_init(struct dept_map *m, struct dept_key *k, int sub,
+ const char *n)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ if(DEPT_WARN_ON(sub < 0 || sub >= DEPT_MAX_SUBCLASSES_USR)) {
+ m->nocheck = true;
+ goto exit;
+ }
+
+ if (m->keys != k)
+ m->keys = k;
+ clean_classes_cache(&m->keys_local);
+
+ m->sub_usr = sub;
+ m->name = n;
+ m->wgen = 0U;
+ m->nocheck = false;
+exit:
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_map_init);
+
+void dept_map_reinit(struct dept_map *m)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ clean_classes_cache(&m->keys_local);
+ m->wgen = 0U;
+ m->nocheck = false;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_map_reinit);
+
+void dept_map_nocheck(struct dept_map *m)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ m->nocheck = true;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_map_nocheck);
+
+static LIST_HEAD(classes);
+
+static inline bool within(const void *addr, void *start, unsigned long size)
+{
+ return addr >= start && addr < start + size;
+}
+
+static void free_staleie_staleiw_range(void *start, unsigned int sz)
+{
+ struct dept_staleie *sie, *sien;
+ struct dept_staleiw *siw, *siwn;
+
+ list_for_each_entry_safe(sie, sien, &staleies, all_node)
+ if (within((void *)sie->ip, start, sz))
+ hash_del_staleie(sie);
+
+ list_for_each_entry_safe(siw, siwn, &staleiws, all_node)
+ if (within((void *)siw->ip, start, sz))
+ hash_del_staleiw(siw);
+}
+
+void dept_free_range(void *start, unsigned int sz)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_class *c, *n;
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * dept_free_range() should not fail.
+ *
+ * FIXME: Should be fixed if dept_free_range() causes deadlock
+ * with dept_lock().
+ */
+ while (unlikely(!dept_lock()));
+
+ /*
+ * Free all staleies and staleiws within the range.
+ */
+ free_staleie_staleiw_range(start, sz);
+
+ list_for_each_entry_safe(c, n, &classes, all_node) {
+ if (!within((void *)c->key, start, sz) &&
+ !within(c->name, start, sz))
+ continue;
+
+ hash_del_class(c);
+ disconnect_class(c);
+ list_del(&c->all_node);
+ invalidate_class(c);
+
+ /*
+ * Actual deletion will happen on the rcu callback
+ * that has been added in disconnect_class().
+ */
+ del_class(c);
+ }
+ dept_unlock();
+ dept_exit(flags);
+
+ /*
+ * Wait until even lockless hash_lookup_class() for the class
+ * returns NULL.
+ */
+ might_sleep();
+ synchronize_rcu();
+}
+
+static inline int map_sub(struct dept_map *m, int e)
+{
+ return m->sub_usr + e * DEPT_MAX_SUBCLASSES_USR;
+}
+
+static struct dept_class *check_new_class(struct dept_key *local,
+ struct dept_key *k, int sub,
+ const char *n)
+{
+ struct dept_class *c = NULL;
+
+ if (DEPT_WARN_ON(sub >= DEPT_MAX_SUBCLASSES))
+ return NULL;
+
+ if (DEPT_WARN_ON(!k))
+ return NULL;
+
+ if (sub < DEPT_MAX_SUBCLASSES_CACHE)
+ c = READ_ONCE(local->classes[sub]);
+
+ if (c)
+ return c;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (c)
+ goto caching;
+
+ if (unlikely(!dept_lock()))
+ return NULL;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (unlikely(c))
+ goto unlock;
+
+ c = new_class();
+ if (unlikely(!c))
+ goto unlock;
+
+ c->name = n;
+ c->sub = sub;
+ c->key = (unsigned long)(k->subkeys + sub);
+ hash_add_class(c);
+ list_add(&c->all_node, &classes);
+unlock:
+ dept_unlock();
+caching:
+ if (sub < DEPT_MAX_SUBCLASSES_CACHE && c)
+ WRITE_ONCE(local->classes[sub], c);
+
+ return c;
+}
+
+void dept_wait(struct dept_map *m, unsigned long w_f, unsigned long ip,
+ const char *w_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int e;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * Be as conservative as possible. In case of mulitple waits for
+ * a single dept_map, we are going to keep only the last wait's
+ * wgen for simplicity - keeping all wgens seems overengineering.
+ *
+ * Of course, it might cause missing some dependencies that
+ * would rarely, probabily never, happen but it helps avoid
+ * false positive report.
+ */
+ for_each_set_bit(e, &w_f, DEPT_MAX_SUBCLASSES_EVT) {
+ struct dept_class *c;
+ struct dept_key *k;
+
+ k = m->keys ?: &m->keys_local;
+ c = check_new_class(&m->keys_local, k,
+ map_sub(m, e), m->name);
+ if (!c)
+ continue;
+
+ add_wait(c, ip, w_fn, ne);
+ }
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_wait);
+
+void dept_ecxt_enter(struct dept_map *m, unsigned long e_f, unsigned long ip,
+ const char *c_fn, const char *e_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int e;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ for_each_set_bit(e, &e_f, DEPT_MAX_SUBCLASSES_EVT) {
+ struct dept_class *c;
+ struct dept_key *k;
+
+ k = m->keys ?: &m->keys_local;
+ c = check_new_class(&m->keys_local, k,
+ map_sub(m, e), m->name);
+ if (!c)
+ continue;
+
+ add_ecxt((void *)m, c, ip, c_fn, e_fn, ne);
+ }
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_ecxt_enter);
+
+void dept_asked_event(struct dept_map *m)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ unsigned int wg;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * Avoid zero wgen.
+ */
+ wg = atomic_inc_return(&wgen) ?: atomic_inc_return(&wgen);
+ WRITE_ONCE(m->wgen, wg);
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_asked_event);
+
+void dept_event(struct dept_map *m, unsigned long e_f, unsigned long ip,
+ const char *e_fn)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int e;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ for_each_set_bit(e, &e_f, DEPT_MAX_SUBCLASSES_EVT) {
+ struct dept_class *c;
+ struct dept_key *k;
+
+ k = m->keys ?: &m->keys_local;
+ c = check_new_class(&m->keys_local, k,
+ map_sub(m, e), m->name);
+ if (!c)
+ continue;
+
+ add_ecxt((void *)m, c, 0UL, NULL, e_fn, 0);
+ do_event((void *)m, c, READ_ONCE(m->wgen), ip);
+ pop_ecxt((void *)m);
+ }
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_event);
+
+void dept_ecxt_exit(struct dept_map *m, unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+ pop_ecxt((void *)m);
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_ecxt_exit);
+
+struct dept_map *dept_top_map(void)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_map *m;
+ unsigned long flags;
+ int pos;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return NULL;
+
+ flags = dept_enter();
+ pos = dt->ecxt_held_pos;
+ m = pos ? (struct dept_map *)dt->ecxt_held[pos - 1].key : NULL;
+ dept_exit(flags);
+
+ return m;
+}
+EXPORT_SYMBOL_GPL(dept_top_map);
+
+void dept_warn_on(bool cond)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+ DEPT_WARN_ON(cond);
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_warn_on);
+
+void dept_task_exit(struct task_struct *t)
+{
+ struct dept_task *dt = &t->dept_task;
+ int i;
+
+ raw_local_irq_disable();
+
+ if (dt->stack)
+ put_stack(dt->stack);
+
+ for (i = 0; i < dt->ecxt_held_pos; i++)
+ put_ecxt(dt->ecxt_held[i].ecxt);
+
+ for (i = 0; i < DEPT_MAX_WAIT_HIST; i++)
+ if (dt->wait_hist[i].wait)
+ put_wait(dt->wait_hist[i].wait);
+
+ dept_off();
+
+ raw_local_irq_enable();
+}
+
+void dept_task_init(struct task_struct *t)
+{
+ memset(&t->dept_task, 0x0, sizeof(struct dept_task));
+}
+
+void dept_key_init(struct dept_key *k)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int sub;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * dept_key_init() should not fail.
+ *
+ * FIXME: Should be fixed if dept_key_init() causes deadlock
+ * with dept_lock().
+ */
+ while (unlikely(!dept_lock()));
+
+ for (sub = 0; sub < DEPT_MAX_SUBCLASSES; sub++) {
+ struct dept_class *c;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (!c)
+ continue;
+
+ DEPT_STOP("The class(%s/%d) has not been removed.\n",
+ c->name, sub);
+ break;
+ }
+
+ dept_unlock();
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_key_init);
+
+void dept_key_destroy(struct dept_key *k)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int sub;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * dept_key_destroy() should not fail.
+ *
+ * FIXME: Should be fixed if dept_key_destroy() causes deadlock
+ * with dept_lock().
+ */
+ while (unlikely(!dept_lock()));
+
+ for (sub = 0; sub < DEPT_MAX_SUBCLASSES; sub++) {
+ struct dept_class *c;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (!c)
+ continue;
+
+ hash_del_class(c);
+ disconnect_class(c);
+ list_del(&c->all_node);
+ invalidate_class(c);
+
+ /*
+ * Actual deletion will happen on the rcu callback
+ * that has been added in disconnect_class().
+ */
+ del_class(c);
+ }
+
+ dept_unlock();
+ dept_exit(flags);
+
+ /*
+ * Wait until even lockless hash_lookup_class() for the class
+ * returns NULL.
+ */
+ might_sleep();
+ synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(dept_key_destroy);
+
+static void move_llist(struct llist_head *to, struct llist_head *from)
+{
+ struct llist_node *first = llist_del_all(from);
+ struct llist_node *last;
+
+ if (!first)
+ return;
+
+ for (last = first; last->next; last = last->next);
+ llist_add_batch(first, last, to);
+}
+
+static void migrate_per_cpu_pool(void)
+{
+ const int boot_cpu = 0;
+ int i;
+
+ /*
+ * The boot CPU has been using the temperal local pool so far.
+ * From now on that per_cpu areas have been ready, use the
+ * per_cpu local pool instead.
+ */
+ DEPT_WARN_ON(smp_processor_id() != boot_cpu);
+ for (i = 0; i < OBJECT_NR; i++) {
+ struct llist_head *from;
+ struct llist_head *to;
+
+ from = &pool[i].boot_pool;
+ to = per_cpu_ptr(pool[i].lpool, boot_cpu);
+ move_llist(to, from);
+ }
+}
+
+#define B2KB(B) ((B) / 1024)
+
+/*
+ * Should be called after setup_per_cpu_areas() and before no non-boot
+ * CPUs have been on.
+ */
+void __init dept_init(void)
+{
+ size_t mem_total = 0;
+
+ local_irq_disable();
+ dept_per_cpu_ready = 1;
+ migrate_per_cpu_pool();
+ local_irq_enable();
+
+#define OBJECT(id, nr) mem_total += sizeof(struct dept_##id) * nr;
+ #include "dept_object.h"
+#undef OBJECT
+#define HASH(id, bits) mem_total += sizeof(struct hlist_head) * (1UL << bits);
+ #include "dept_hash.h"
+#undef HASH
+
+ printk("DEPendency Tracker: Copyright (c) 2020 LG Electronics, Inc., Byungchul Park\n");
+ printk("... DEPT_MAX_STACK_ENTRY: %d\n", DEPT_MAX_STACK_ENTRY);
+ printk("... DEPT_MAX_WAIT_HIST : %d\n", DEPT_MAX_WAIT_HIST );
+ printk("... DEPT_MAX_ECXT_HELD : %d\n", DEPT_MAX_ECXT_HELD );
+ printk("... DEPT_MAX_SUBCLASSES : %d\n", DEPT_MAX_SUBCLASSES );
+#define OBJECT(id, nr) \
+ printk("... memory used by %s: %zu KB\n", \
+ #id, B2KB(sizeof(struct dept_##id) * nr));
+ #include "dept_object.h"
+#undef OBJECT
+#define HASH(id, bits) \
+ printk("... hash list head used by %s: %zu KB\n", \
+ #id, B2KB(sizeof(struct hlist_head) * (1UL << bits)));
+ #include "dept_hash.h"
+#undef HASH
+ printk("... total memory used by objects and hashs: %zu KB\n", B2KB(mem_total));
+ printk("... per task memory footprint: %zu bytes\n", sizeof(struct dept_task));
+}
diff --git a/kernel/dependency/dept_hash.h b/kernel/dependency/dept_hash.h
new file mode 100644
index 0000000..4762977
--- /dev/null
+++ b/kernel/dependency/dept_hash.h
@@ -0,0 +1,11 @@
+/*
+ * HASH(id, bits)
+ *
+ * id : Id for the object of struct dept_##id.
+ * bits: 1UL << bits is the hash table size.
+ */
+
+HASH(dep , 12)
+HASH(class , 12)
+HASH(staleiw , 4)
+HASH(staleie , 4)
diff --git a/kernel/dependency/dept_object.h b/kernel/dependency/dept_object.h
new file mode 100644
index 0000000..3e8ba81
--- /dev/null
+++ b/kernel/dependency/dept_object.h
@@ -0,0 +1,14 @@
+/*
+ * OBJECT(id, nr)
+ *
+ * id: Id for the object of struct dept_##id.
+ * nr: # of the object that should be kept in the pool.
+ */
+
+OBJECT(dep ,1024 * 8)
+OBJECT(class ,1024 * 4)
+OBJECT(stack ,1024 * 32)
+OBJECT(ecxt ,1024 * 4)
+OBJECT(wait ,1024 * 32)
+OBJECT(staleiw ,16)
+OBJECT(staleie ,16)
diff --git a/kernel/exit.c b/kernel/exit.c
index 1f236ed..cb4c026 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -857,6 +857,7 @@ void __noreturn do_exit(long code)
exit_tasks_rcu_finish();
lockdep_free_task(tsk);
+ dept_task_exit(tsk);
do_task_dead();
}
EXPORT_SYMBOL_GPL(do_exit);
diff --git a/kernel/fork.c b/kernel/fork.c
index 6d26638..9c22152 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -96,6 +96,7 @@
#include <linux/kasan.h>
#include <linux/scs.h>
#include <linux/io_uring.h>
+#include <linux/dept.h>
#include <asm/pgalloc.h>
#include <linux/uaccess.h>
@@ -2052,6 +2053,7 @@ static __latent_entropy struct task_struct *copy_process(
#ifdef CONFIG_LOCKDEP
lockdep_init_task(p);
#endif
+ dept_task_init(p);
#ifdef CONFIG_DEBUG_MUTEXES
p->blocked_on = NULL; /* not blocked yet */
diff --git a/kernel/module.c b/kernel/module.c
index a4fa44a..7911f32c 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -2258,6 +2258,7 @@ static void free_module(struct module *mod)
/* Free lock-classes; relies on the preceding sync_rcu(). */
lockdep_free_key_range(mod->core_layout.base, mod->core_layout.size);
+ dept_free_range(mod->core_layout.base, mod->core_layout.size);
/* Finally, free the core (containing the module structure) */
module_memfree(mod->core_layout.base);
@@ -4019,6 +4020,7 @@ static int load_module(struct load_info *info, const char __user *uargs,
free_module:
/* Free lock-classes; relies on the preceding sync_rcu() */
lockdep_free_key_range(mod->core_layout.base, mod->core_layout.size);
+ dept_free_range(mod->core_layout.base, mod->core_layout.size);
module_deallocate(mod, info);
free_copy:
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 09229ad..a94567b 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -132,7 +132,7 @@ void __local_bh_disable_ip(unsigned long ip, unsigned int cnt)
* Were softirqs turned off above:
*/
if (softirq_count() == (cnt & SOFTIRQ_MASK))
- lockdep_softirqs_off(ip);
+ trace_softirqs_off_caller(ip);
raw_local_irq_restore(flags);
if (preempt_count() == cnt) {
@@ -153,7 +153,7 @@ static void __local_bh_enable(unsigned int cnt)
trace_preempt_on(CALLER_ADDR0, get_lock_parent_ip());
if (softirq_count() == (cnt & SOFTIRQ_MASK))
- lockdep_softirqs_on(_RET_IP_);
+ trace_softirqs_on_caller(_RET_IP_);
__preempt_count_sub(cnt);
}
@@ -180,7 +180,7 @@ void __local_bh_enable_ip(unsigned long ip, unsigned int cnt)
* Are softirqs going to be turned on now:
*/
if (softirq_count() == SOFTIRQ_DISABLE_OFFSET)
- lockdep_softirqs_on(ip);
+ trace_softirqs_on_caller(ip);
/*
* Keep preemption disabled until we are done with
* softirq processing:
diff --git a/kernel/trace/trace_preemptirq.c b/kernel/trace/trace_preemptirq.c
index f493804..19cafdfb 100644
--- a/kernel/trace/trace_preemptirq.c
+++ b/kernel/trace/trace_preemptirq.c
@@ -19,6 +19,18 @@
/* Per-cpu variable to prevent redundant calls when IRQs already off */
static DEFINE_PER_CPU(int, tracing_irq_cpu);
+void trace_softirqs_on_caller(unsigned long ip)
+{
+ lockdep_softirqs_on(ip);
+ dept_enable_softirq(ip);
+}
+
+void trace_softirqs_off_caller(unsigned long ip)
+{
+ lockdep_softirqs_off(ip);
+ dept_disable_softirq(ip);
+}
+
/*
* Like trace_hardirqs_on() but without the lockdep invocation. This is
* used in the low level entry code where the ordering vs. RCU is important
@@ -33,6 +45,7 @@ void trace_hardirqs_on_prepare(void)
tracer_hardirqs_on(CALLER_ADDR0, CALLER_ADDR1);
this_cpu_write(tracing_irq_cpu, 0);
}
+ dept_enable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_on_prepare);
NOKPROBE_SYMBOL(trace_hardirqs_on_prepare);
@@ -45,6 +58,7 @@ void trace_hardirqs_on(void)
tracer_hardirqs_on(CALLER_ADDR0, CALLER_ADDR1);
this_cpu_write(tracing_irq_cpu, 0);
}
+ dept_enable_hardirq(CALLER_ADDR0);
lockdep_hardirqs_on_prepare(CALLER_ADDR0);
lockdep_hardirqs_on(CALLER_ADDR0);
@@ -66,7 +80,7 @@ void trace_hardirqs_off_finish(void)
if (!in_nmi())
trace_irq_disable(CALLER_ADDR0, CALLER_ADDR1);
}
-
+ dept_disable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_off_finish);
NOKPROBE_SYMBOL(trace_hardirqs_off_finish);
@@ -81,6 +95,7 @@ void trace_hardirqs_off(void)
if (!in_nmi())
trace_irq_disable_rcuidle(CALLER_ADDR0, CALLER_ADDR1);
}
+ dept_disable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_off);
NOKPROBE_SYMBOL(trace_hardirqs_off);
@@ -93,6 +108,7 @@ __visible void trace_hardirqs_on_caller(unsigned long caller_addr)
tracer_hardirqs_on(CALLER_ADDR0, caller_addr);
this_cpu_write(tracing_irq_cpu, 0);
}
+ dept_enable_hardirq(CALLER_ADDR0);
lockdep_hardirqs_on_prepare(CALLER_ADDR0);
lockdep_hardirqs_on(CALLER_ADDR0);
@@ -110,6 +126,7 @@ __visible void trace_hardirqs_off_caller(unsigned long caller_addr)
if (!in_nmi())
trace_irq_disable_rcuidle(CALLER_ADDR0, caller_addr);
}
+ dept_disable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_off_caller);
NOKPROBE_SYMBOL(trace_hardirqs_off_caller);
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index c789b39..51cb690 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1144,6 +1144,26 @@ config DEBUG_PREEMPT
menu "Lock Debugging (spinlocks, mutexes, etc...)"
+config DEPT
+ bool "Dependency tracking"
+ depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
+ select DEBUG_SPINLOCK
+ select DEBUG_MUTEXES
+ select DEBUG_RT_MUTEXES if RT_MUTEXES
+ select DEBUG_RWSEMS
+ select DEBUG_WW_MUTEX_SLOWPATH
+ select DEBUG_LOCK_ALLOC
+ select TRACE_IRQFLAGS
+ select STACKTRACE
+ select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86
+ select KALLSYMS
+ select KALLSYMS_ALL
+ default n
+ help
+ Check dependencies between wait and event and report it if
+ deadlock possibility has been detected. Multiple reports are
+ allowed if there are more than a single problem.
+
config LOCK_DEBUGGING_SUPPORT
bool
depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 03/14] dept: Embed Dept data in Lockdep
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
2022-01-25 2:36 ` [RFC 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
2022-01-25 2:36 ` [RFC 02/14] dept: Implement Dept(Dependency Tracker) Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 04/14] dept: Apply Dept to spinlock Byungchul Park
` (12 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Dept should work independently from Lockdep. However, there's no choise
but to rely on Lockdep code and its instances for now.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 71 ++++++++++++++++++++++++++++++++++++++++---
include/linux/lockdep_types.h | 3 ++
kernel/locking/lockdep.c | 12 ++++----
3 files changed, 76 insertions(+), 10 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index f559487..bcd9d7d 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -20,6 +20,33 @@
extern int prove_locking;
extern int lock_stat;
+#ifdef CONFIG_DEPT
+static inline void dept_after_copy_map(struct dept_map *to,
+ struct dept_map *from)
+{
+ int i;
+
+ if (from->keys == &from->keys_local)
+ to->keys = &to->keys_local;
+
+ if (!to->keys)
+ return;
+
+ /*
+ * Since the class cache can be modified concurrently we could observe
+ * half pointers (64bit arch using 32bit copy insns). Therefore clear
+ * the caches and take the performance hit.
+ *
+ * XXX it doesn't work well with lockdep_set_class_and_subclass(), since
+ * that relies on cache abuse.
+ */
+ for (i = 0; i < DEPT_MAX_SUBCLASSES_CACHE; i++)
+ to->keys->classes[i] = NULL;
+}
+#else
+#define dept_after_copy_map(t, f) do { } while (0)
+#endif
+
#ifdef CONFIG_LOCKDEP
#include <linux/linkage.h>
@@ -43,6 +70,8 @@ static inline void lockdep_copy_map(struct lockdep_map *to,
*/
for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++)
to->class_cache[i] = NULL;
+
+ dept_after_copy_map(&to->dmap, &from->dmap);
}
/*
@@ -176,8 +205,19 @@ struct held_lock {
current->lockdep_recursion -= LOCKDEP_OFF; \
} while (0)
-extern void lockdep_register_key(struct lock_class_key *key);
-extern void lockdep_unregister_key(struct lock_class_key *key);
+extern void __lockdep_register_key(struct lock_class_key *key);
+extern void __lockdep_unregister_key(struct lock_class_key *key);
+
+#define lockdep_register_key(k) \
+do { \
+ __lockdep_register_key(k); \
+ dept_key_init(&(k)->dkey); \
+} while (0)
+#define lockdep_unregister_key(k) \
+do { \
+ __lockdep_unregister_key(k); \
+ dept_key_destroy(&(k)->dkey); \
+} while (0)
/*
* These methods are used by specific locking variants (spinlocks,
@@ -185,9 +225,18 @@ struct held_lock {
* to lockdep:
*/
-extern void lockdep_init_map_waits(struct lockdep_map *lock, const char *name,
+extern void __lockdep_init_map_waits(struct lockdep_map *lock, const char *name,
struct lock_class_key *key, int subclass, short inner, short outer);
+#define lockdep_init_map_waits(l, n, k, s, i, o) \
+do { \
+ __lockdep_init_map_waits(l, n, k, s, i, o); \
+ if ((k) == &__lockdep_no_validate__) \
+ dept_map_nocheck(&(l)->dmap); \
+ else \
+ dept_map_init(&(l)->dmap, &(k)->dkey, s, n); \
+} while (0)
+
static inline void
lockdep_init_map_wait(struct lockdep_map *lock, const char *name,
struct lock_class_key *key, int subclass, short inner)
@@ -398,13 +447,27 @@ enum xhlock_context_t {
XHLOCK_CTX_NR,
};
+#ifdef CONFIG_DEPT
+/*
+ * TODO: I found the case to use an address of other than a real key as
+ * _key, for instance, in workqueue. So for now, we cannot use the
+ * assignment like '.dmap.keys = &(_key)->dkey' unless it's fixed.
+ */
+#define STATIC_DEPT_MAP_INIT(_name, _key) .dmap = { \
+ .name = (_name), \
+ .keys = NULL },
+#else
+#define STATIC_DEPT_MAP_INIT(_name, _key)
+#endif
+
#define lockdep_init_map_crosslock(m, n, k, s) do {} while (0)
/*
* To initialize a lockdep_map statically use this macro.
* Note that _name must not be NULL.
*/
#define STATIC_LOCKDEP_MAP_INIT(_name, _key) \
- { .name = (_name), .key = (void *)(_key), }
+ { .name = (_name), .key = (void *)(_key), \
+ STATIC_DEPT_MAP_INIT(_name, _key) }
static inline void lockdep_invariant_state(bool force) {}
static inline void lockdep_free_task(struct task_struct *task) {}
diff --git a/include/linux/lockdep_types.h b/include/linux/lockdep_types.h
index 9a1fd49..822ab1d 100644
--- a/include/linux/lockdep_types.h
+++ b/include/linux/lockdep_types.h
@@ -11,6 +11,7 @@
#define __LINUX_LOCKDEP_TYPES_H
#include <linux/types.h>
+#include <linux/dept.h>
#define MAX_LOCKDEP_SUBCLASSES 8UL
@@ -70,6 +71,7 @@ struct lock_class_key {
struct hlist_node hash_entry;
struct lockdep_subclass_key subkeys[MAX_LOCKDEP_SUBCLASSES];
};
+ struct dept_key dkey;
};
extern struct lock_class_key __lockdep_no_validate__;
@@ -175,6 +177,7 @@ struct lockdep_map {
int cpu;
unsigned long ip;
#endif
+ struct dept_map dmap;
};
struct pin_cookie { unsigned int val; };
diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index c1418b47..b649588 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -1168,7 +1168,7 @@ static inline struct hlist_head *keyhashentry(const struct lock_class_key *key)
}
/* Register a dynamically allocated key. */
-void lockdep_register_key(struct lock_class_key *key)
+void __lockdep_register_key(struct lock_class_key *key)
{
struct hlist_head *hash_head;
struct lock_class_key *k;
@@ -1191,7 +1191,7 @@ void lockdep_register_key(struct lock_class_key *key)
restore_irqs:
raw_local_irq_restore(flags);
}
-EXPORT_SYMBOL_GPL(lockdep_register_key);
+EXPORT_SYMBOL_GPL(__lockdep_register_key);
/* Check whether a key has been registered as a dynamic key. */
static bool is_dynamic_key(const struct lock_class_key *key)
@@ -4577,7 +4577,7 @@ static inline int check_wait_context(struct task_struct *curr,
/*
* Initialize a lock instance's lock-class mapping info:
*/
-void lockdep_init_map_waits(struct lockdep_map *lock, const char *name,
+void __lockdep_init_map_waits(struct lockdep_map *lock, const char *name,
struct lock_class_key *key, int subclass,
short inner, short outer)
{
@@ -4636,7 +4636,7 @@ void lockdep_init_map_waits(struct lockdep_map *lock, const char *name,
raw_local_irq_restore(flags);
}
}
-EXPORT_SYMBOL_GPL(lockdep_init_map_waits);
+EXPORT_SYMBOL_GPL(__lockdep_init_map_waits);
struct lock_class_key __lockdep_no_validate__;
EXPORT_SYMBOL_GPL(__lockdep_no_validate__);
@@ -6085,7 +6085,7 @@ void lockdep_reset_lock(struct lockdep_map *lock)
}
/* Unregister a dynamically allocated key. */
-void lockdep_unregister_key(struct lock_class_key *key)
+void __lockdep_unregister_key(struct lock_class_key *key)
{
struct hlist_head *hash_head = keyhashentry(key);
struct lock_class_key *k;
@@ -6120,7 +6120,7 @@ void lockdep_unregister_key(struct lock_class_key *key)
/* Wait until is_dynamic_key() has finished accessing k->hash_entry. */
synchronize_rcu();
}
-EXPORT_SYMBOL_GPL(lockdep_unregister_key);
+EXPORT_SYMBOL_GPL(__lockdep_unregister_key);
void __init lockdep_init(void)
{
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 04/14] dept: Apply Dept to spinlock
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (2 preceding siblings ...)
2022-01-25 2:36 ` [RFC 03/14] dept: Embed Dept data in Lockdep Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 05/14] dept: Apply Dept to mutex families Byungchul Park
` (11 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes Dept able to track dependencies by spinlock.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 18 +++++++++++++++---
include/linux/spinlock.h | 24 ++++++++++++++++++++++++
include/linux/spinlock_types.h | 10 ++++++++++
3 files changed, 49 insertions(+), 3 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index bcd9d7d..c08e3ae 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -566,9 +566,21 @@ static inline void print_irqtrace_events(struct task_struct *curr)
#define lock_acquire_shared(l, s, t, n, i) lock_acquire(l, s, t, 1, 1, n, i)
#define lock_acquire_shared_recursive(l, s, t, n, i) lock_acquire(l, s, t, 2, 1, n, i)
-#define spin_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
-#define spin_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
-#define spin_release(l, i) lock_release(l, i)
+#define spin_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_spin_lock(&(l)->dmap, s, t, NULL, "spin_unlock", i); \
+} while (0)
+#define spin_acquire_nest(l, s, t, n, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, n, i); \
+ dept_spin_lock(&(l)->dmap, s, t, (n) ? &(n)->dmap : NULL, "spin_unlock", i); \
+} while (0)
+#define spin_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_spin_unlock(&(l)->dmap, i); \
+} while (0)
#define rwlock_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
#define rwlock_acquire_read(l, s, t, i) \
diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index 7989784..b25bfc1 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -92,6 +92,30 @@
# include <linux/spinlock_up.h>
#endif
+#ifdef CONFIG_DEPT
+#define dept_spin_lock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, 1UL, ip, __func__, ne); \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_spin_unlock(m, ip) \
+do { \
+ dept_event(m, 1UL, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_spin_lock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_spin_unlock(m, ip) do { } while (0)
+#endif
+
#ifdef CONFIG_DEBUG_SPINLOCK
extern void __raw_spin_lock_init(raw_spinlock_t *lock, const char *name,
struct lock_class_key *key, short inner);
diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h
index b981caa..4ed8a9b 100644
--- a/include/linux/spinlock_types.h
+++ b/include/linux/spinlock_types.h
@@ -32,16 +32,26 @@
#define SPINLOCK_OWNER_INIT ((void *)-1L)
+#ifdef CONFIG_DEPT
+# define RAW_SPIN_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+# define SPIN_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define RAW_SPIN_DMAP_INIT(lockname)
+# define SPIN_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define RAW_SPIN_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_SPIN, \
+ RAW_SPIN_DMAP_INIT(lockname) \
}
# define SPIN_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_CONFIG, \
+ SPIN_DMAP_INIT(lockname) \
}
#else
# define RAW_SPIN_DEP_MAP_INIT(lockname)
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 05/14] dept: Apply Dept to mutex families
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (3 preceding siblings ...)
2022-01-25 2:36 ` [RFC 04/14] dept: Apply Dept to spinlock Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 06/14] dept: Apply Dept to rwlock Byungchul Park
` (10 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes Dept able to track dependencies by mutex families.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 18 +++++++++++++++---
include/linux/mutex.h | 31 +++++++++++++++++++++++++++++++
include/linux/rtmutex.h | 11 ++++++++++-
3 files changed, 56 insertions(+), 4 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index c08e3ae..f558091 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -597,9 +597,21 @@ static inline void print_irqtrace_events(struct task_struct *curr)
#define seqcount_acquire_read(l, s, t, i) lock_acquire_shared_recursive(l, s, t, NULL, i)
#define seqcount_release(l, i) lock_release(l, i)
-#define mutex_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
-#define mutex_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
-#define mutex_release(l, i) lock_release(l, i)
+#define mutex_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_mutex_lock(&(l)->dmap, s, t, NULL, "mutex_unlock", i); \
+} while (0)
+#define mutex_acquire_nest(l, s, t, n, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, n, i); \
+ dept_mutex_lock(&(l)->dmap, s, t, (n) ? &(n)->dmap : NULL, "mutex_unlock", i);\
+} while (0)
+#define mutex_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_mutex_unlock(&(l)->dmap, i); \
+} while (0)
#define rwsem_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
#define rwsem_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
diff --git a/include/linux/mutex.h b/include/linux/mutex.h
index dcd185c..a3477e7 100644
--- a/include/linux/mutex.h
+++ b/include/linux/mutex.h
@@ -76,6 +76,30 @@ struct ww_mutex {
#endif
};
+#ifdef CONFIG_DEPT
+#define dept_mutex_lock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, 1UL, ip, __func__, ne); \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_mutex_unlock(m, ip) \
+do { \
+ dept_event(m, 1UL, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_mutex_lock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_mutex_unlock(m, ip) do { } while (0)
+#endif
+
/*
* This is the control structure for tasks blocked on mutex,
* which resides on the blocked task's kernel stack:
@@ -119,11 +143,18 @@ static inline void mutex_destroy(struct mutex *lock) {}
__mutex_init((mutex), #mutex, &__key); \
} while (0)
+#ifdef CONFIG_DEPT
+# define DMAP_MUTEX_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define DMAP_MUTEX_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define __DEP_MAP_MUTEX_INITIALIZER(lockname) \
, .dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_SLEEP, \
+ DMAP_MUTEX_INIT(lockname) \
}
#else
# define __DEP_MAP_MUTEX_INITIALIZER(lockname)
diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h
index 6fd615a..52d4444 100644
--- a/include/linux/rtmutex.h
+++ b/include/linux/rtmutex.h
@@ -75,9 +75,18 @@ struct rt_mutex {
# define rt_mutex_debug_task_free(t) do { } while (0)
#endif
+#ifdef CONFIG_DEPT
+#define DMAP_RT_MUTEX_INIT(mutexname) .dmap = { .name = #mutexname },
+#else
+#define DMAP_RT_MUTEX_INIT(mutexname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
#define __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname) \
- , .dep_map = { .name = #mutexname }
+ , .dep_map = { \
+ .name = #mutexname, \
+ DMAP_RT_MUTEX_INIT(mutexname) \
+ }
#else
#define __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname)
#endif
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 06/14] dept: Apply Dept to rwlock
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (4 preceding siblings ...)
2022-01-25 2:36 ` [RFC 05/14] dept: Apply Dept to mutex families Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 07/14] dept: Apply Dept to wait_for_completion()/complete() Byungchul Park
` (9 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes Dept able to track dependencies by rwlock.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 25 +++++++++++++++++-----
include/linux/rwlock.h | 48 ++++++++++++++++++++++++++++++++++++++++++
include/linux/rwlock_api_smp.h | 8 +++----
include/linux/rwlock_types.h | 7 ++++++
4 files changed, 79 insertions(+), 9 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index f558091..4653df7 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -582,16 +582,31 @@ static inline void print_irqtrace_events(struct task_struct *curr)
dept_spin_unlock(&(l)->dmap, i); \
} while (0)
-#define rwlock_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
+#define rwlock_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_rwlock_wlock(&(l)->dmap, s, t, NULL, "write_unlock", i); \
+} while (0)
#define rwlock_acquire_read(l, s, t, i) \
do { \
- if (read_lock_is_recursive()) \
+ if (read_lock_is_recursive()) { \
lock_acquire_shared_recursive(l, s, t, NULL, i); \
- else \
+ dept_rwlock_rlock(&(l)->dmap, s, t, NULL, "read_unlock", i, 0);\
+ } else { \
lock_acquire_shared(l, s, t, NULL, i); \
+ dept_rwlock_rlock(&(l)->dmap, s, t, NULL, "read_unlock", i, 1);\
+ } \
+} while (0)
+#define rwlock_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_rwlock_wunlock(&(l)->dmap, i); \
+} while (0)
+#define rwlock_release_read(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_rwlock_runlock(&(l)->dmap, i); \
} while (0)
-
-#define rwlock_release(l, i) lock_release(l, i)
#define seqcount_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
#define seqcount_acquire_read(l, s, t, i) lock_acquire_shared_recursive(l, s, t, NULL, i)
diff --git a/include/linux/rwlock.h b/include/linux/rwlock.h
index 3dcd617..6792521 100644
--- a/include/linux/rwlock.h
+++ b/include/linux/rwlock.h
@@ -28,6 +28,54 @@
do { *(lock) = __RW_LOCK_UNLOCKED(lock); } while (0)
#endif
+#ifdef CONFIG_DEPT
+#define DEPT_EVT_RWLOCK_R 1UL
+#define DEPT_EVT_RWLOCK_W (1UL << 1)
+#define DEPT_EVT_RWLOCK_RW (DEPT_EVT_RWLOCK_R | DEPT_EVT_RWLOCK_W)
+
+#define dept_rwlock_wlock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_W, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, DEPT_EVT_RWLOCK_RW, ip, __func__, ne); \
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_W, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } \
+}while (0)
+#define dept_rwlock_rlock(m, ne, t, n, e_fn, ip, q) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_R, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, (q) ? DEPT_EVT_RWLOCK_RW : DEPT_EVT_RWLOCK_W, ip, __func__, ne);\
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_R, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_rwlock_wunlock(m, ip) \
+do { \
+ dept_event(m, DEPT_EVT_RWLOCK_W, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#define dept_rwlock_runlock(m, ip) \
+do { \
+ dept_event(m, DEPT_EVT_RWLOCK_R, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_rwlock_wlock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_rwlock_rlock(m, ne, t, n, e_fn, ip, q) do { } while (0)
+#define dept_rwlock_wunlock(m, ip) do { } while (0)
+#define dept_rwlock_runlock(m, ip) do { } while (0)
+#endif
+
#ifdef CONFIG_DEBUG_SPINLOCK
extern void do_raw_read_lock(rwlock_t *lock) __acquires(lock);
#define do_raw_read_lock_flags(lock, flags) do_raw_read_lock(lock)
diff --git a/include/linux/rwlock_api_smp.h b/include/linux/rwlock_api_smp.h
index abfb53a..9f47df9 100644
--- a/include/linux/rwlock_api_smp.h
+++ b/include/linux/rwlock_api_smp.h
@@ -222,7 +222,7 @@ static inline void __raw_write_unlock(rwlock_t *lock)
static inline void __raw_read_unlock(rwlock_t *lock)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
preempt_enable();
}
@@ -230,7 +230,7 @@ static inline void __raw_read_unlock(rwlock_t *lock)
static inline void
__raw_read_unlock_irqrestore(rwlock_t *lock, unsigned long flags)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
local_irq_restore(flags);
preempt_enable();
@@ -238,7 +238,7 @@ static inline void __raw_read_unlock(rwlock_t *lock)
static inline void __raw_read_unlock_irq(rwlock_t *lock)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
local_irq_enable();
preempt_enable();
@@ -246,7 +246,7 @@ static inline void __raw_read_unlock_irq(rwlock_t *lock)
static inline void __raw_read_unlock_bh(rwlock_t *lock)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
__local_bh_enable_ip(_RET_IP_, SOFTIRQ_LOCK_OFFSET);
}
diff --git a/include/linux/rwlock_types.h b/include/linux/rwlock_types.h
index 3bd03e1..a366f39 100644
--- a/include/linux/rwlock_types.h
+++ b/include/linux/rwlock_types.h
@@ -21,11 +21,18 @@
#define RWLOCK_MAGIC 0xdeaf1eed
+#ifdef CONFIG_DEPT
+# define RW_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define RW_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define RW_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_CONFIG, \
+ RW_DMAP_INIT(lockname) \
}
#else
# define RW_DEP_MAP_INIT(lockname)
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 07/14] dept: Apply Dept to wait_for_completion()/complete()
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (5 preceding siblings ...)
2022-01-25 2:36 ` [RFC 06/14] dept: Apply Dept to rwlock Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 08/14] dept: Apply Dept to seqlock Byungchul Park
` (8 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes Dept able to track dependencies by
wait_for_completion()/complete().
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/completion.h | 48 ++++++++++++++++++++++++++++++++++++++++++----
kernel/sched/completion.c | 12 ++++++++++--
2 files changed, 54 insertions(+), 6 deletions(-)
diff --git a/include/linux/completion.h b/include/linux/completion.h
index bf8e770..7f4a200 100644
--- a/include/linux/completion.h
+++ b/include/linux/completion.h
@@ -26,15 +26,51 @@
struct completion {
unsigned int done;
struct swait_queue_head wait;
+ struct dept_map dmap;
};
-#define init_completion_map(x, m) __init_completion(x)
-#define init_completion(x) __init_completion(x)
+#ifdef CONFIG_DEPT
+#define dept_wfc_init(m, k, s, n) dept_map_init(m, k, s, n)
+#define dept_wfc_reinit(m) dept_map_reinit(m)
+#define dept_wfc_wait(m, ip) \
+do { \
+ dept_asked_event(m); \
+ dept_wait(m, 1UL, ip, __func__, 0); \
+} while (0)
+#define dept_wfc_complete(m, ip) dept_event(m, 1UL, ip, __func__)
+#define dept_wfc_enter(m, ip) dept_ecxt_enter(m, 1UL, ip, "completion_context_enter", "complete", 0)
+#define dept_wfc_exit(m, ip) dept_ecxt_exit(m, ip)
+#else
+#define dept_wfc_init(m, k, s, n) do { (void)(n); (void)(k); } while (0)
+#define dept_wfc_reinit(m) do { } while (0)
+#define dept_wfc_wait(m, ip) do { } while (0)
+#define dept_wfc_complete(m, ip) do { } while (0)
+#define dept_wfc_enter(m, ip) do { } while (0)
+#define dept_wfc_exit(m, ip) do { } while (0)
+#endif
+
+#ifdef CONFIG_DEPT
+#define WFC_DEPT_MAP_INIT(work) .dmap = { .name = #work }
+#else
+#define WFC_DEPT_MAP_INIT(work)
+#endif
+
+#define init_completion_map(x, m) \
+ do { \
+ static struct dept_key __dkey; \
+ __init_completion(x, &__dkey, #x); \
+ } while (0)
+#define init_completion(x) \
+ do { \
+ static struct dept_key __dkey; \
+ __init_completion(x, &__dkey, #x); \
+ } while (0)
static inline void complete_acquire(struct completion *x) {}
static inline void complete_release(struct completion *x) {}
#define COMPLETION_INITIALIZER(work) \
- { 0, __SWAIT_QUEUE_HEAD_INITIALIZER((work).wait) }
+ { 0, __SWAIT_QUEUE_HEAD_INITIALIZER((work).wait), \
+ WFC_DEPT_MAP_INIT(work) }
#define COMPLETION_INITIALIZER_ONSTACK_MAP(work, map) \
(*({ init_completion_map(&(work), &(map)); &(work); }))
@@ -82,9 +118,12 @@ static inline void complete_release(struct completion *x) {}
* This inline function will initialize a dynamically created completion
* structure.
*/
-static inline void __init_completion(struct completion *x)
+static inline void __init_completion(struct completion *x,
+ struct dept_key *dkey,
+ const char *name)
{
x->done = 0;
+ dept_wfc_init(&x->dmap, dkey, 0, name);
init_swait_queue_head(&x->wait);
}
@@ -98,6 +137,7 @@ static inline void __init_completion(struct completion *x)
static inline void reinit_completion(struct completion *x)
{
x->done = 0;
+ dept_wfc_reinit(&x->dmap);
}
extern void wait_for_completion(struct completion *);
diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c
index a778554..6e31cc0 100644
--- a/kernel/sched/completion.c
+++ b/kernel/sched/completion.c
@@ -29,6 +29,7 @@ void complete(struct completion *x)
{
unsigned long flags;
+ dept_wfc_complete(&x->dmap, _RET_IP_);
raw_spin_lock_irqsave(&x->wait.lock, flags);
if (x->done != UINT_MAX)
@@ -58,6 +59,7 @@ void complete_all(struct completion *x)
{
unsigned long flags;
+ dept_wfc_complete(&x->dmap, _RET_IP_);
lockdep_assert_RT_in_threaded_ctx();
raw_spin_lock_irqsave(&x->wait.lock, flags);
@@ -112,17 +114,23 @@ void complete_all(struct completion *x)
}
static long __sched
-wait_for_common(struct completion *x, long timeout, int state)
+_wait_for_common(struct completion *x, long timeout, int state)
{
return __wait_for_common(x, schedule_timeout, timeout, state);
}
static long __sched
-wait_for_common_io(struct completion *x, long timeout, int state)
+_wait_for_common_io(struct completion *x, long timeout, int state)
{
return __wait_for_common(x, io_schedule_timeout, timeout, state);
}
+#define wait_for_common(x, t, s) \
+({ dept_wfc_wait(&(x)->dmap, _RET_IP_); _wait_for_common(x, t, s); })
+
+#define wait_for_common_io(x, t, s) \
+({ dept_wfc_wait(&(x)->dmap, _RET_IP_); _wait_for_common_io(x, t, s); })
+
/**
* wait_for_completion: - waits for completion of a task
* @x: holds the state of this particular completion
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 08/14] dept: Apply Dept to seqlock
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (6 preceding siblings ...)
2022-01-25 2:36 ` [RFC 07/14] dept: Apply Dept to wait_for_completion()/complete() Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 09/14] dept: Apply Dept to rwsem Byungchul Park
` (7 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes Dept able to track dependencies by seqlock with adding wait
annotation on read side of seqlock.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/seqlock.h | 19 ++++++++++++++++++-
1 file changed, 18 insertions(+), 1 deletion(-)
diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h
index cbfc78b..622267f 100644
--- a/include/linux/seqlock.h
+++ b/include/linux/seqlock.h
@@ -23,6 +23,13 @@
#include <asm/processor.h>
+#ifdef CONFIG_DEPT
+#define DEPT_EVT_ALL ((1UL << DEPT_MAX_SUBCLASSES_EVT) - 1)
+#define dept_seq_wait(m, ip) dept_wait(m, DEPT_EVT_ALL, ip, __func__, 0)
+#else
+#define dept_seq_wait(m, ip) do { } while (0)
+#endif
+
/*
* The seqlock seqcount_t interface does not prescribe a precise sequence of
* read begin/retry/end. For readers, typically there is a call to
@@ -148,7 +155,7 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s)
* This lock-unlock technique must be implemented for all of PREEMPT_RT
* sleeping locks. See Documentation/locking/locktypes.rst
*/
-#if defined(CONFIG_LOCKDEP) || defined(CONFIG_PREEMPT_RT)
+#if defined(CONFIG_LOCKDEP) || defined(CONFIG_DEPT) || defined(CONFIG_PREEMPT_RT)
#define __SEQ_LOCK(expr) expr
#else
#define __SEQ_LOCK(expr)
@@ -203,6 +210,12 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s)
__SEQ_LOCK(locktype *lock); \
} seqcount_##lockname##_t; \
\
+static __always_inline void \
+__seqprop_##lockname##_wait(const seqcount_##lockname##_t *s) \
+{ \
+ __SEQ_LOCK(dept_seq_wait(&(lockmember)->dep_map.dmap, _RET_IP_));\
+} \
+ \
static __always_inline seqcount_t * \
__seqprop_##lockname##_ptr(seqcount_##lockname##_t *s) \
{ \
@@ -271,6 +284,8 @@ static inline void __seqprop_assert(const seqcount_t *s)
lockdep_assert_preemption_disabled();
}
+static inline void __seqprop_wait(const seqcount_t *s) { }
+
#define __SEQ_RT IS_ENABLED(CONFIG_PREEMPT_RT)
SEQCOUNT_LOCKNAME(raw_spinlock, raw_spinlock_t, false, s->lock, raw_spin, raw_spin_lock(s->lock))
@@ -311,6 +326,7 @@ static inline void __seqprop_assert(const seqcount_t *s)
#define __seqcount_sequence(s) __seqprop(s, sequence)
#define __seqcount_lock_preemptible(s) __seqprop(s, preemptible)
#define __seqcount_assert_lock_held(s) __seqprop(s, assert)
+#define __seqcount_dept_wait(s) __seqprop(s, wait)
/**
* __read_seqcount_begin() - begin a seqcount_t read section w/o barrier
@@ -360,6 +376,7 @@ static inline void __seqprop_assert(const seqcount_t *s)
#define read_seqcount_begin(s) \
({ \
seqcount_lockdep_reader_access(__seqcount_ptr(s)); \
+ __seqcount_dept_wait(s); \
raw_read_seqcount_begin(s); \
})
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 09/14] dept: Apply Dept to rwsem
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (7 preceding siblings ...)
2022-01-25 2:36 ` [RFC 08/14] dept: Apply Dept to seqlock Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 10/14] dept: Add proc knobs to show stats and dependency graph Byungchul Park
` (6 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes Dept able to track dependencies by rwsem.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 24 ++++++++++++++++++++----
include/linux/percpu-rwsem.h | 10 +++++++++-
include/linux/rwsem.h | 31 +++++++++++++++++++++++++++++++
3 files changed, 60 insertions(+), 5 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 4653df7..80e665e 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -628,10 +628,26 @@ static inline void print_irqtrace_events(struct task_struct *curr)
dept_mutex_unlock(&(l)->dmap, i); \
} while (0)
-#define rwsem_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
-#define rwsem_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
-#define rwsem_acquire_read(l, s, t, i) lock_acquire_shared(l, s, t, NULL, i)
-#define rwsem_release(l, i) lock_release(l, i)
+#define rwsem_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_rwsem_lock(&(l)->dmap, s, t, NULL, "up_write", i); \
+} while (0)
+#define rwsem_acquire_nest(l, s, t, n, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, n, i); \
+ dept_rwsem_lock(&(l)->dmap, s, t, (n) ? &(n)->dmap : NULL, "up_write", i);\
+} while (0)
+#define rwsem_acquire_read(l, s, t, i) \
+do { \
+ lock_acquire_shared(l, s, t, NULL, i); \
+ dept_rwsem_lock(&(l)->dmap, s, t, NULL, "up_read", i); \
+} while (0)
+#define rwsem_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_rwsem_unlock(&(l)->dmap, i); \
+} while (0)
#define lock_map_acquire(l) lock_acquire_exclusive(l, 0, 0, NULL, _THIS_IP_)
#define lock_map_acquire_read(l) lock_acquire_shared_recursive(l, 0, 0, NULL, _THIS_IP_)
diff --git a/include/linux/percpu-rwsem.h b/include/linux/percpu-rwsem.h
index 5fda40f..7ec5625 100644
--- a/include/linux/percpu-rwsem.h
+++ b/include/linux/percpu-rwsem.h
@@ -20,8 +20,16 @@ struct percpu_rw_semaphore {
#endif
};
+#ifdef CONFIG_DEPT
+#define __PERCPU_RWSEM_DMAP_INIT(lockname) .dmap = { .name = #lockname }
+#else
+#define __PERCPU_RWSEM_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
-#define __PERCPU_RWSEM_DEP_MAP_INIT(lockname) .dep_map = { .name = #lockname },
+#define __PERCPU_RWSEM_DEP_MAP_INIT(lockname) .dep_map = { \
+ .name = #lockname, \
+ __PERCPU_RWSEM_DMAP_INIT(lockname) },
#else
#define __PERCPU_RWSEM_DEP_MAP_INIT(lockname)
#endif
diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h
index 25e3fde8..b4a266f 100644
--- a/include/linux/rwsem.h
+++ b/include/linux/rwsem.h
@@ -20,6 +20,30 @@
#include <linux/osq_lock.h>
#endif
+#ifdef CONFIG_DEPT
+#define dept_rwsem_lock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, 1UL, ip, __func__, ne); \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_rwsem_unlock(m, ip) \
+do { \
+ dept_event(m, 1UL, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_rwsem_lock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_rwsem_unlock(m, ip) do { } while (0)
+#endif
+
/*
* For an uncontended rwsem, count and owner are the only fields a task
* needs to touch when acquiring the rwsem. So they are put next to each
@@ -64,11 +88,18 @@ static inline int rwsem_is_locked(struct rw_semaphore *sem)
/* Common initializer macros and functions */
+#ifdef CONFIG_DEPT
+# define RWSEM_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define RWSEM_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define __RWSEM_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_SLEEP, \
+ RWSEM_DMAP_INIT(lockname) \
},
#else
# define __RWSEM_DEP_MAP_INIT(lockname)
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 10/14] dept: Add proc knobs to show stats and dependency graph
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (8 preceding siblings ...)
2022-01-25 2:36 ` [RFC 09/14] dept: Apply Dept to rwsem Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 11/14] dept: Introduce split map concept and new APIs for them Byungchul Park
` (5 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
It'd be useful to show Dept internal stats and dependency graph on
runtime via proc for better information. Introduced the knobs.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
kernel/dependency/Makefile | 1 +
kernel/dependency/dept.c | 24 ++++------
kernel/dependency/dept_internal.h | 26 +++++++++++
kernel/dependency/dept_proc.c | 97 +++++++++++++++++++++++++++++++++++++++
4 files changed, 133 insertions(+), 15 deletions(-)
create mode 100644 kernel/dependency/dept_internal.h
create mode 100644 kernel/dependency/dept_proc.c
diff --git a/kernel/dependency/Makefile b/kernel/dependency/Makefile
index 9f7778e..49152ad 100644
--- a/kernel/dependency/Makefile
+++ b/kernel/dependency/Makefile
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: GPL-2.0
obj-$(CONFIG_DEPT) += dept.o
+obj-$(CONFIG_DEPT) += dept_proc.o
diff --git a/kernel/dependency/dept.c b/kernel/dependency/dept.c
index 64287894..b682bf9 100644
--- a/kernel/dependency/dept.c
+++ b/kernel/dependency/dept.c
@@ -90,6 +90,7 @@
#include <linux/hash.h>
#include <linux/dept.h>
#include <linux/utsname.h>
+#include "dept_internal.h"
static int dept_stop;
static int dept_per_cpu_ready;
@@ -241,20 +242,13 @@ static inline struct dept_task *dept_task(void)
* have been freed will be placed.
*/
-enum object_t {
-#define OBJECT(id, nr) OBJECT_##id,
- #include "dept_object.h"
-#undef OBJECT
- OBJECT_NR,
-};
-
#define OBJECT(id, nr) \
static struct dept_##id spool_##id[nr]; \
static DEFINE_PER_CPU(struct llist_head, lpool_##id);
#include "dept_object.h"
#undef OBJECT
-static struct dept_pool pool[OBJECT_NR] = {
+struct dept_pool dept_pool[OBJECT_NR] = {
#define OBJECT(id, nr) { \
.name = #id, \
.obj_sz = sizeof(struct dept_##id), \
@@ -283,7 +277,7 @@ static void *from_pool(enum object_t t)
if (DEPT_WARN_ON(!irqs_disabled()))
return NULL;
- p = &pool[t];
+ p = &dept_pool[t];
/*
* Try local pool first.
@@ -312,7 +306,7 @@ static void *from_pool(enum object_t t)
static void to_pool(void *o, enum object_t t)
{
- struct dept_pool *p = &pool[t];
+ struct dept_pool *p = &dept_pool[t];
struct llist_head *h;
preempt_disable();
@@ -1986,7 +1980,7 @@ void dept_map_nocheck(struct dept_map *m)
}
EXPORT_SYMBOL_GPL(dept_map_nocheck);
-static LIST_HEAD(classes);
+LIST_HEAD(dept_classes);
static inline bool within(const void *addr, void *start, unsigned long size)
{
@@ -2031,7 +2025,7 @@ void dept_free_range(void *start, unsigned int sz)
*/
free_staleie_staleiw_range(start, sz);
- list_for_each_entry_safe(c, n, &classes, all_node) {
+ list_for_each_entry_safe(c, n, &dept_classes, all_node) {
if (!within((void *)c->key, start, sz) &&
!within(c->name, start, sz))
continue;
@@ -2100,7 +2094,7 @@ static struct dept_class *check_new_class(struct dept_key *local,
c->sub = sub;
c->key = (unsigned long)(k->subkeys + sub);
hash_add_class(c);
- list_add(&c->all_node, &classes);
+ list_add(&c->all_node, &dept_classes);
unlock:
dept_unlock();
caching:
@@ -2429,8 +2423,8 @@ static void migrate_per_cpu_pool(void)
struct llist_head *from;
struct llist_head *to;
- from = &pool[i].boot_pool;
- to = per_cpu_ptr(pool[i].lpool, boot_cpu);
+ from = &dept_pool[i].boot_pool;
+ to = per_cpu_ptr(dept_pool[i].lpool, boot_cpu);
move_llist(to, from);
}
}
diff --git a/kernel/dependency/dept_internal.h b/kernel/dependency/dept_internal.h
new file mode 100644
index 0000000..007c1ee
--- /dev/null
+++ b/kernel/dependency/dept_internal.h
@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept(DEPendency Tracker) - runtime dependency tracker internal header
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ */
+
+#ifndef __DEPT_INTERNAL_H
+#define __DEPT_INTERNAL_H
+
+#ifdef CONFIG_DEPT
+
+enum object_t {
+#define OBJECT(id, nr) OBJECT_##id,
+ #include "dept_object.h"
+#undef OBJECT
+ OBJECT_NR,
+};
+
+extern struct list_head dept_classes;
+extern struct dept_pool dept_pool[];
+
+#endif
+#endif /* __DEPT_INTERNAL_H */
diff --git a/kernel/dependency/dept_proc.c b/kernel/dependency/dept_proc.c
new file mode 100644
index 0000000..5e5afa3
--- /dev/null
+++ b/kernel/dependency/dept_proc.c
@@ -0,0 +1,97 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * kernel/dependency/dept_proc.c
+ *
+ * Runtime dependency tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (C) 2021 LG Electronics, Inc. , Byungchul Park
+ *
+ * Code for /proc/dept_*:
+ *
+ */
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/dept.h>
+#include "dept_internal.h"
+
+static void *l_next(struct seq_file *m, void *v, loff_t *pos)
+{
+ /*
+ * XXX: Serialize list traversal if needed. The following might
+ * give a wrong information on contention.
+ */
+ return seq_list_next(v, &dept_classes, pos);
+}
+
+static void *l_start(struct seq_file *m, loff_t *pos)
+{
+ /*
+ * XXX: Serialize list traversal if needed. The following might
+ * give a wrong information on contention.
+ */
+ return seq_list_start_head(&dept_classes, *pos);
+}
+
+static void l_stop(struct seq_file *m, void *v)
+{
+}
+
+static int l_show(struct seq_file *m, void *v)
+{
+ struct dept_class *fc = list_entry(v, struct dept_class, all_node);
+ struct dept_dep *d;
+
+ if (v == &dept_classes) {
+ seq_printf(m, "All classes:\n\n");
+ return 0;
+ }
+
+ seq_printf(m, "[%p] %s\n", (void *)fc->key, fc->name);
+
+ /*
+ * XXX: Serialize list traversal if needed. The following might
+ * give a wrong information on contention.
+ */
+ list_for_each_entry(d, &fc->dep_head, dep_node) {
+ struct dept_class *tc = d->wait->class;
+ seq_printf(m, " -> [%p] %s\n", (void *)tc->key, tc->name);
+ }
+ seq_puts(m, "\n");
+
+ return 0;
+}
+
+static const struct seq_operations dept_deps_ops = {
+ .start = l_start,
+ .next = l_next,
+ .stop = l_stop,
+ .show = l_show,
+};
+
+static int dept_stats_show(struct seq_file *m, void *v)
+{
+ int r;
+
+ seq_printf(m, "Availablity in the static pools:\n\n");
+#define OBJECT(id, nr) \
+ r = atomic_read(&dept_pool[OBJECT_##id].obj_nr); \
+ if (r < 0) \
+ r = 0; \
+ seq_printf(m, "%s\t%d/%d(%d%%)\n", #id, r, nr, (r * 100) / (nr));
+ #include "dept_object.h"
+#undef OBJECT
+
+ return 0;
+}
+
+static int __init dept_proc_init(void)
+{
+ proc_create_seq("dept_deps", S_IRUSR, NULL, &dept_deps_ops);
+ proc_create_single("dept_stats", S_IRUSR, NULL, dept_stats_show);
+ return 0;
+}
+
+__initcall(dept_proc_init);
+
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 11/14] dept: Introduce split map concept and new APIs for them
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (9 preceding siblings ...)
2022-01-25 2:36 ` [RFC 10/14] dept: Add proc knobs to show stats and dependency graph Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback} Byungchul Park
` (4 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
There is a case where all maps used for a type of wait/event is so large
in size. For instance, struct page can be a type for (un)lock_page().
The additional memory size for the maps would be 'the # of pages *
sizeof(struct dept_map)' if each struct page keeps its map all the way,
which might be too big to accept in some system.
It'd better to have splitted map, one is for each instance and the other
is for the type which is commonly used, and new APIs using them. So
introduced split map and new APIs for them.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept.h | 36 ++++++++++++++
kernel/dependency/dept.c | 122 +++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 158 insertions(+)
diff --git a/include/linux/dept.h b/include/linux/dept.h
index 8665629..5a726bf 100644
--- a/include/linux/dept.h
+++ b/include/linux/dept.h
@@ -441,6 +441,30 @@ struct dept_map {
bool nocheck;
};
+struct dept_map_each {
+ /*
+ * wait timestamp associated to this map
+ */
+ unsigned int wgen;
+};
+
+struct dept_map_common {
+ const char *name;
+ struct dept_key *keys;
+ int sub_usr;
+
+ /*
+ * It's local copy for fast acces to the associated classes. And
+ * Also used for dept_key instance for statically defined map.
+ */
+ struct dept_key keys_local;
+
+ /*
+ * whether this map should be going to be checked or not
+ */
+ bool nocheck;
+};
+
#define DEPT_TASK_INITIALIZER(t) \
.dept_task.wait_hist = { { .wait = NULL, } }, \
.dept_task.ecxt_held_pos = 0, \
@@ -469,6 +493,11 @@ struct dept_map {
extern void dept_ecxt_exit(struct dept_map *m, unsigned long ip);
extern struct dept_map *dept_top_map(void);
extern void dept_warn_on(bool cond);
+extern void dept_split_map_each_init(struct dept_map_each *me);
+extern void dept_split_map_common_init(struct dept_map_common *mc, struct dept_key *k, const char *n);
+extern void dept_wait_split_map(struct dept_map_each *me, struct dept_map_common *mc, unsigned long ip, const char *w_fn, int ne);
+extern void dept_event_split_map(struct dept_map_each *me, struct dept_map_common *mc, unsigned long ip, const char *e_fn);
+extern void dept_asked_event_split_map(struct dept_map_each *me, struct dept_map_common *mc);
/*
* for users who want to manage external keys
@@ -505,6 +534,8 @@ struct dept_map {
struct dept_task { };
struct dept_key { };
struct dept_map { };
+struct dept_map_each { };
+struct dept_map_commmon { };
#define DEPT_TASK_INITIALIZER(t)
@@ -525,6 +556,11 @@ struct dept_map {
#define dept_ecxt_exit(m, ip) do { } while (0)
#define dept_top_map() NULL
#define dept_warn_on(c) do { } while (0)
+#define dept_split_map_each_init(me) do { } while (0)
+#define dept_split_map_common_init(mc, k, n) do { (void)(n); (void)(k); } while (0)
+#define dept_wait_split_map(me, mc, ip, w_fn, ne) do { } while (0)
+#define dept_event_split_map(me, mc, ip, e_fn) do { } while (0)
+#define dept_asked_event_split_map(me, mc) do { } while (0)
#define dept_key_init(k) do { (void)(k); } while (0)
#define dept_key_destroy(k) do { (void)(k); } while (0)
diff --git a/kernel/dependency/dept.c b/kernel/dependency/dept.c
index b682bf9..bc764cc 100644
--- a/kernel/dependency/dept.c
+++ b/kernel/dependency/dept.c
@@ -2233,6 +2233,128 @@ void dept_event(struct dept_map *m, unsigned long e_f, unsigned long ip,
}
EXPORT_SYMBOL_GPL(dept_event);
+void dept_split_map_each_init(struct dept_map_each *me)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ me->wgen = 0U;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_split_map_each_init);
+
+void dept_split_map_common_init(struct dept_map_common *mc,
+ struct dept_key *k, const char *n)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ if (mc->keys != k)
+ mc->keys = k;
+ clean_classes_cache(&mc->keys_local);
+
+ /*
+ * sub_usr is not used with split map.
+ */
+ mc->sub_usr = 0;
+ mc->name = n;
+ mc->nocheck = false;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_split_map_common_init);
+
+void dept_wait_split_map(struct dept_map_each *me,
+ struct dept_map_common *mc,
+ unsigned long ip, const char *w_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_class *c;
+ struct dept_key *k;
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (mc->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ k = mc->keys ?: &mc->keys_local;
+ c = check_new_class(&mc->keys_local, k, 0, mc->name);
+ if (c)
+ add_wait(c, ip, w_fn, ne);
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_wait_split_map);
+
+void dept_asked_event_split_map(struct dept_map_each *me,
+ struct dept_map_common *mc)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ unsigned int wg;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (mc->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * Avoid zero wgen.
+ */
+ wg = atomic_inc_return(&wgen) ?: atomic_inc_return(&wgen);
+ WRITE_ONCE(me->wgen, wg);
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_asked_event_split_map);
+
+void dept_event_split_map(struct dept_map_each *me,
+ struct dept_map_common *mc,
+ unsigned long ip, const char *e_fn)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_class *c;
+ struct dept_key *k;
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (mc->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ k = mc->keys ?: &mc->keys_local;
+ c = check_new_class(&mc->keys_local, k, 0, mc->name);
+ if (c) {
+ add_ecxt((void *)me, c, 0UL, NULL, e_fn, 0);
+ do_event((void *)me, c, READ_ONCE(me->wgen), ip);
+ pop_ecxt((void *)me);
+ }
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_event_split_map);
+
void dept_ecxt_exit(struct dept_map *m, unsigned long ip)
{
struct dept_task *dt = dept_task();
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback}
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (10 preceding siblings ...)
2022-01-25 2:36 ` [RFC 11/14] dept: Introduce split map concept and new APIs for them Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header Byungchul Park
` (3 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes Dept able to track dependencies by PG_{locked,writeback}. For
instance, (un)lock_page() generates that type of dependency.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept_page.h | 71 +++++++++++++++++++++++++++++++++++++++++
include/linux/page-flags.h | 26 +++++++++++++--
include/linux/pagemap.h | 7 +++-
init/main.c | 2 ++
kernel/dependency/dept_object.h | 2 +-
lib/Kconfig.debug | 1 +
mm/filemap.c | 62 +++++++++++++++++++++++++++++++++++
mm/page_ext.c | 5 +++
8 files changed, 172 insertions(+), 4 deletions(-)
create mode 100644 include/linux/dept_page.h
diff --git a/include/linux/dept_page.h b/include/linux/dept_page.h
new file mode 100644
index 0000000..3fa75d2
--- /dev/null
+++ b/include/linux/dept_page.h
@@ -0,0 +1,71 @@
+#ifndef __LINUX_DEPT_PAGE_H
+#define __LINUX_DEPT_PAGE_H
+
+#ifdef CONFIG_DEPT
+#include <linux/dept.h>
+
+extern struct page_ext_operations dept_pglocked_ops;
+extern struct page_ext_operations dept_pgwriteback_ops;
+extern struct dept_map_common pglocked_mc;
+extern struct dept_map_common pgwriteback_mc;
+
+extern void dept_page_init(void);
+extern struct dept_map_each *get_pglocked_me(struct page *page);
+extern struct dept_map_each *get_pgwriteback_me(struct page *page);
+
+#define dept_pglocked_wait(p) \
+do { \
+ struct dept_map_each *me = get_pglocked_me(p); \
+ if (likely(me)) \
+ dept_wait_split_map(me, &pglocked_mc, _RET_IP_, \
+ __func__, 0); \
+} while (0)
+
+#define dept_pglocked_set_bit(p) \
+do { \
+ struct dept_map_each *me = get_pglocked_me(p); \
+ if (likely(me)) \
+ dept_asked_event_split_map(me, &pglocked_mc); \
+} while (0)
+
+#define dept_pglocked_event(p) \
+do { \
+ struct dept_map_each *me = get_pglocked_me(p); \
+ if (likely(me)) \
+ dept_event_split_map(me, &pglocked_mc, _RET_IP_,\
+ __func__); \
+} while (0)
+
+#define dept_pgwriteback_wait(p) \
+do { \
+ struct dept_map_each *me = get_pgwriteback_me(p); \
+ if (likely(me)) \
+ dept_wait_split_map(me, &pgwriteback_mc, _RET_IP_,\
+ __func__, 0); \
+} while (0)
+
+#define dept_pgwriteback_set_bit(p) \
+do { \
+ struct dept_map_each *me = get_pgwriteback_me(p); \
+ if (likely(me)) \
+ dept_asked_event_split_map(me, &pgwriteback_mc);\
+} while (0)
+
+#define dept_pgwriteback_event(p) \
+do { \
+ struct dept_map_each *me = get_pgwriteback_me(p); \
+ if (likely(me)) \
+ dept_event_split_map(me, &pgwriteback_mc, _RET_IP_,\
+ __func__); \
+} while (0)
+#else
+#define dept_page_init() do { } while (0)
+#define dept_pglocked_wait(p) do { } while (0)
+#define dept_pglocked_set_bit(p) do { } while (0)
+#define dept_pglocked_event(p) do { } while (0)
+#define dept_pgwriteback_wait(p) do { } while (0)
+#define dept_pgwriteback_set_bit(p) do { } while (0)
+#define dept_pgwriteback_event(p) do { } while (0)
+#endif
+
+#endif /* __LINUX_DEPT_PAGE_H */
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 4f6ba93..4e79f4e 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -326,7 +326,6 @@ static inline void page_init_poison(struct page *page, size_t size)
#define TESTSCFLAG_FALSE(uname) \
TESTSETFLAG_FALSE(uname) TESTCLEARFLAG_FALSE(uname)
-__PAGEFLAG(Locked, locked, PF_NO_TAIL)
PAGEFLAG(Waiters, waiters, PF_ONLY_HEAD) __CLEARPAGEFLAG(Waiters, waiters, PF_ONLY_HEAD)
PAGEFLAG(Error, error, PF_NO_TAIL) TESTCLEARFLAG(Error, error, PF_NO_TAIL)
PAGEFLAG(Referenced, referenced, PF_HEAD)
@@ -374,7 +373,6 @@ static inline void page_init_poison(struct page *page, size_t size)
* risky: they bypass page accounting.
*/
TESTPAGEFLAG(Writeback, writeback, PF_NO_TAIL)
- TESTSCFLAG(Writeback, writeback, PF_NO_TAIL)
PAGEFLAG(MappedToDisk, mappedtodisk, PF_NO_TAIL)
/* PG_readahead is only used for reads; PG_reclaim is only for writes */
@@ -444,6 +442,30 @@ static __always_inline int PageSwapCache(struct page *page)
PAGEFLAG(Idle, idle, PF_ANY)
#endif
+#ifdef CONFIG_DEPT
+TESTPAGEFLAG(Locked, locked, PF_NO_TAIL)
+__CLEARPAGEFLAG(Locked, locked, PF_NO_TAIL)
+TESTCLEARFLAG(Writeback, writeback, PF_NO_TAIL)
+
+#include <linux/dept_page.h>
+static __always_inline void __SetPageLocked(struct page *page)
+{
+ dept_pglocked_set_bit(page);
+ __set_bit(PG_locked, &PF_NO_TAIL(page, 1)->flags);
+}
+
+static __always_inline int TestSetPageWriteback(struct page *page)
+{
+ int ret = test_and_set_bit(PG_writeback, &PF_NO_TAIL(page, 1)->flags);
+ if (!ret)
+ dept_pgwriteback_set_bit(page);
+ return ret;
+}
+#else
+__PAGEFLAG(Locked, locked, PF_NO_TAIL)
+TESTSCFLAG(Writeback, writeback, PF_NO_TAIL)
+#endif
+
/*
* PageReported() is used to track reported free pages within the Buddy
* allocator. We can use the non-atomic version of the test and set
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index d5570de..475a880 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -15,6 +15,7 @@
#include <linux/bitops.h>
#include <linux/hardirq.h> /* for in_interrupt() */
#include <linux/hugetlb_inline.h>
+#include <linux/dept_page.h>
struct pagevec;
@@ -597,8 +598,12 @@ extern int __lock_page_or_retry(struct page *page, struct mm_struct *mm,
*/
static inline int trylock_page(struct page *page)
{
+ int ret;
page = compound_head(page);
- return (likely(!test_and_set_bit_lock(PG_locked, &page->flags)));
+ ret = test_and_set_bit_lock(PG_locked, &page->flags);
+ if (likely(!ret))
+ dept_pglocked_set_bit(page);
+ return likely(!ret);
}
/*
diff --git a/init/main.c b/init/main.c
index 1818a90..5a6f474 100644
--- a/init/main.c
+++ b/init/main.c
@@ -99,6 +99,7 @@
#include <linux/mem_encrypt.h>
#include <linux/kcsan.h>
#include <linux/init_syscalls.h>
+#include <linux/pagemap.h>
#include <asm/io.h>
#include <asm/bugs.h>
@@ -993,6 +994,7 @@ asmlinkage __visible void __init __no_sanitize_address start_kernel(void)
lockdep_init();
dept_init();
+ dept_page_init();
/*
* Need to run this when irqs are enabled, because it wants
diff --git a/kernel/dependency/dept_object.h b/kernel/dependency/dept_object.h
index 3e8ba81..2c4f5f0 100644
--- a/kernel/dependency/dept_object.h
+++ b/kernel/dependency/dept_object.h
@@ -5,7 +5,7 @@
* nr: # of the object that should be kept in the pool.
*/
-OBJECT(dep ,1024 * 8)
+OBJECT(dep ,1024 * 16)
OBJECT(class ,1024 * 4)
OBJECT(stack ,1024 * 32)
OBJECT(ecxt ,1024 * 4)
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 51cb690..fd88c3f 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1153,6 +1153,7 @@ config DEPT
select DEBUG_RWSEMS
select DEBUG_WW_MUTEX_SLOWPATH
select DEBUG_LOCK_ALLOC
+ select PAGE_EXTENSION
select TRACE_IRQFLAGS
select STACKTRACE
select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86
diff --git a/mm/filemap.c b/mm/filemap.c
index 0b2067b..20cbdc4 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1100,6 +1100,11 @@ static void wake_up_page_bit(struct page *page, int bit_nr)
unsigned long flags;
wait_queue_entry_t bookmark;
+ if (bit_nr == PG_locked)
+ dept_pglocked_event(page);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_event(page);
+
key.page = page;
key.bit_nr = bit_nr;
key.page_match = 0;
@@ -1179,6 +1184,10 @@ static inline bool trylock_page_bit_common(struct page *page, int bit_nr,
if (wait->flags & WQ_FLAG_EXCLUSIVE) {
if (test_and_set_bit(bit_nr, &page->flags))
return false;
+ else if (bit_nr == PG_locked)
+ dept_pglocked_set_bit(page);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_set_bit(page);
} else if (test_bit(bit_nr, &page->flags))
return false;
@@ -1199,6 +1208,11 @@ static inline int wait_on_page_bit_common(wait_queue_head_t *q,
bool delayacct = false;
unsigned long pflags;
+ if (bit_nr == PG_locked)
+ dept_pglocked_wait(page);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_wait(page);
+
if (bit_nr == PG_locked &&
!PageUptodate(page) && PageWorkingset(page)) {
if (!PageSwapBacked(page)) {
@@ -1291,6 +1305,11 @@ static inline int wait_on_page_bit_common(wait_queue_head_t *q,
if (unlikely(test_and_set_bit(bit_nr, &page->flags)))
goto repeat;
+ if (bit_nr == PG_locked)
+ dept_pglocked_set_bit(page);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_set_bit(page);
+
wait->flags |= WQ_FLAG_DONE;
break;
}
@@ -3527,3 +3546,46 @@ int try_to_release_page(struct page *page, gfp_t gfp_mask)
}
EXPORT_SYMBOL(try_to_release_page);
+
+#ifdef CONFIG_DEPT
+static bool need_dept_pglocked(void)
+{
+ return true;
+}
+
+struct page_ext_operations dept_pglocked_ops = {
+ .size = sizeof(struct dept_map_each),
+ .need = need_dept_pglocked,
+};
+
+struct dept_map_each *get_pglocked_me(struct page *p)
+{
+ struct page_ext *e = lookup_page_ext(p);
+ return e ? (void *)e + dept_pglocked_ops.offset : NULL;
+}
+
+static bool need_dept_pgwriteback(void)
+{
+ return true;
+}
+
+struct page_ext_operations dept_pgwriteback_ops = {
+ .size = sizeof(struct dept_map_each),
+ .need = need_dept_pgwriteback,
+};
+
+struct dept_map_each *get_pgwriteback_me(struct page *p)
+{
+ struct page_ext *e = lookup_page_ext(p);
+ return e ? (void *)e + dept_pgwriteback_ops.offset : NULL;
+}
+
+struct dept_map_common pglocked_mc;
+struct dept_map_common pgwriteback_mc;
+
+void dept_page_init(void)
+{
+ dept_split_map_common_init(&pglocked_mc, NULL, "pglocked");
+ dept_split_map_common_init(&pgwriteback_mc, NULL, "pgwriteback");
+}
+#endif
diff --git a/mm/page_ext.c b/mm/page_ext.c
index a3616f7..0126931 100644
--- a/mm/page_ext.c
+++ b/mm/page_ext.c
@@ -8,6 +8,7 @@
#include <linux/kmemleak.h>
#include <linux/page_owner.h>
#include <linux/page_idle.h>
+#include <linux/dept_page.h>
/*
* struct page extension
@@ -65,6 +66,10 @@
#if defined(CONFIG_IDLE_PAGE_TRACKING) && !defined(CONFIG_64BIT)
&page_idle_ops,
#endif
+#ifdef CONFIG_DEPT
+ &dept_pglocked_ops,
+ &dept_pgwriteback_ops,
+#endif
};
unsigned long page_ext_size = sizeof(struct page_ext);
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (11 preceding siblings ...)
2022-01-25 2:36 ` [RFC 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback} Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-25 2:36 ` [RFC 14/14] dept: Apply SDT to swait Byungchul Park
` (2 subsequent siblings)
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Now that Dept has already been applied to major synchronization
machanisms e.g. spinlock, wait_for_completion and the like, we can take
advantage of Dept by default if CONFIG_DEPT is on.
However, sometimes we need to manually tag wait/event on places where
Dept hasn't been applied. SDT(Single-event Dependency Tracker) would be
useful in that case to tag those. The usage is like:
1. Initialize dmap in the instance associated to the wait/event.
sdt_map_init(dmap);
2. Add the following just before the wait.
sdt_wait(dmap);
3. Add the following just before the event.
sdt_event(dmap);
For better reference of SDT APIs, separated the APIs from dept.h.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept.h | 36 --------------------------------
include/linux/dept_sdt.h | 53 ++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 53 insertions(+), 36 deletions(-)
create mode 100644 include/linux/dept_sdt.h
diff --git a/include/linux/dept.h b/include/linux/dept.h
index 5a726bf..10ef7fc 100644
--- a/include/linux/dept.h
+++ b/include/linux/dept.h
@@ -504,32 +504,6 @@ struct dept_map_common {
*/
extern void dept_key_init(struct dept_key *k);
extern void dept_key_destroy(struct dept_key *k);
-
-#define DEPT_SDT_MAP_INIT(dname) { .name = #dname }
-#define DEFINE_DEPT_SDT(x) \
- struct dept_map x = DEPT_SDT_MAP_INIT(x)
-
-/*
- * SDT(Single-event Dependency Tracker) APIs
- *
- * In case that one dept_map instance maps to a single event, SDT APIs
- * can be used.
- */
-#define sdt_map_init(m) \
- do { \
- static struct dept_key __key; \
- dept_map_init(m, &__key, 0, #m); \
- } while (0)
-#define sdt_map_init_key(m, k) dept_map_init(m, k, 0, #m)
-
-#define sdt_wait(m) \
- do { \
- dept_asked_event(m); \
- dept_wait(m, 1UL, _THIS_IP_, "wait", 0); \
- } while (0)
-#define sdt_ecxt_enter(m) dept_ecxt_enter(m, 1UL, _THIS_IP_, "start", "event", 0)
-#define sdt_event(m) dept_event(m, 1UL, _THIS_IP_, "event")
-#define sdt_ecxt_exit(m) dept_ecxt_exit(m, _THIS_IP_)
#else /* !CONFIG_DEPT */
struct dept_task { };
struct dept_key { };
@@ -563,15 +537,5 @@ struct dept_map_common {
#define dept_asked_event_split_map(me, mc) do { } while (0)
#define dept_key_init(k) do { (void)(k); } while (0)
#define dept_key_destroy(k) do { (void)(k); } while (0)
-
-#define DEPT_SDT_MAP_INIT(dname)
-#define DEFINE_DEPT_SDT(x)
-
-#define sdt_map_init(m) do { } while (0)
-#define sdt_map_init_key(m, k) do { (void)(k); } while (0)
-#define sdt_wait(m) do { } while (0)
-#define sdt_ecxt_enter(m) do { } while (0)
-#define sdt_event(m) do { } while (0)
-#define sdt_ecxt_exit(m) do { } while (0)
#endif
#endif /* __LINUX_DEPT_H */
diff --git a/include/linux/dept_sdt.h b/include/linux/dept_sdt.h
new file mode 100644
index 0000000..32a4a9e
--- /dev/null
+++ b/include/linux/dept_sdt.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept Single-event Dependency Tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ */
+
+#ifndef __LINUX_DEPT_SDT_H
+#define __LINUX_DEPT_SDT_H
+
+#ifdef CONFIG_DEPT
+
+#include <linux/dept.h>
+
+#define DEPT_SDT_MAP_INIT(dname) { .name = #dname }
+#define DEFINE_DEPT_SDT(x) \
+ struct dept_map x = DEPT_SDT_MAP_INIT(x)
+
+/*
+ * SDT(Single-event Dependency Tracker) APIs
+ *
+ * In case that one dept_map instance maps to a single event, SDT APIs
+ * can be used.
+ */
+#define sdt_map_init(m) \
+ do { \
+ static struct dept_key __key; \
+ dept_map_init(m, &__key, 0, #m); \
+ } while (0)
+#define sdt_map_init_key(m, k) dept_map_init(m, k, 0, #m)
+
+#define sdt_wait(m) \
+ do { \
+ dept_asked_event(m); \
+ dept_wait(m, 1UL, _THIS_IP_, "wait", 0); \
+ } while (0)
+#define sdt_ecxt_enter(m) dept_ecxt_enter(m, 1UL, _THIS_IP_, "start", "event", 0)
+#define sdt_event(m) dept_event(m, 1UL, _THIS_IP_, "event")
+#define sdt_ecxt_exit(m) dept_ecxt_exit(m, _THIS_IP_)
+#else /* !CONFIG_DEPT */
+#define DEPT_SDT_MAP_INIT(dname)
+#define DEFINE_DEPT_SDT(x)
+
+#define sdt_map_init(m) do { } while (0)
+#define sdt_map_init_key(m, k) do { (void)(k); } while (0)
+#define sdt_wait(m) do { } while (0)
+#define sdt_ecxt_enter(m) do { } while (0)
+#define sdt_event(m) do { } while (0)
+#define sdt_ecxt_exit(m) do { } while (0)
+#endif
+#endif /* __LINUX_DEPT_SDT_H */
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [RFC 14/14] dept: Apply SDT to swait
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (12 preceding siblings ...)
2022-01-25 2:36 ` [RFC 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header Byungchul Park
@ 2022-01-25 2:36 ` Byungchul Park
2022-01-27 1:00 ` [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
2022-01-27 1:08 ` Patches(Dept) for v5.17-rc1 Byungchul Park
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-25 2:36 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team
Makes SDT able to track dependencies by swait.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/swait.h | 4 ++++
kernel/sched/swait.c | 8 ++++++++
2 files changed, 12 insertions(+)
diff --git a/include/linux/swait.h b/include/linux/swait.h
index 6a8c22b..dbdf2ce 100644
--- a/include/linux/swait.h
+++ b/include/linux/swait.h
@@ -6,6 +6,7 @@
#include <linux/stddef.h>
#include <linux/spinlock.h>
#include <linux/wait.h>
+#include <linux/dept_sdt.h>
#include <asm/current.h>
/*
@@ -43,6 +44,7 @@
struct swait_queue_head {
raw_spinlock_t lock;
struct list_head task_list;
+ struct dept_map dmap;
};
struct swait_queue {
@@ -61,6 +63,7 @@ struct swait_queue {
#define __SWAIT_QUEUE_HEAD_INITIALIZER(name) { \
.lock = __RAW_SPIN_LOCK_UNLOCKED(name.lock), \
.task_list = LIST_HEAD_INIT((name).task_list), \
+ .dmap = DEPT_SDT_MAP_INIT(name), \
}
#define DECLARE_SWAIT_QUEUE_HEAD(name) \
@@ -72,6 +75,7 @@ extern void __init_swait_queue_head(struct swait_queue_head *q, const char *name
#define init_swait_queue_head(q) \
do { \
static struct lock_class_key __key; \
+ sdt_map_init(&(q)->dmap); \
__init_swait_queue_head((q), #q, &__key); \
} while (0)
diff --git a/kernel/sched/swait.c b/kernel/sched/swait.c
index e1c655f..b6c2efb 100644
--- a/kernel/sched/swait.c
+++ b/kernel/sched/swait.c
@@ -27,6 +27,7 @@ void swake_up_locked(struct swait_queue_head *q)
return;
curr = list_first_entry(&q->task_list, typeof(*curr), task_list);
+ sdt_event(&q->dmap);
wake_up_process(curr->task);
list_del_init(&curr->task_list);
}
@@ -69,6 +70,7 @@ void swake_up_all(struct swait_queue_head *q)
while (!list_empty(&tmp)) {
curr = list_first_entry(&tmp, typeof(*curr), task_list);
+ sdt_event(&q->dmap);
wake_up_state(curr->task, TASK_NORMAL);
list_del_init(&curr->task_list);
@@ -97,6 +99,9 @@ void prepare_to_swait_exclusive(struct swait_queue_head *q, struct swait_queue *
__prepare_to_swait(q, wait);
set_current_state(state);
raw_spin_unlock_irqrestore(&q->lock, flags);
+
+ if (state & TASK_NORMAL)
+ sdt_wait(&q->dmap);
}
EXPORT_SYMBOL(prepare_to_swait_exclusive);
@@ -119,6 +124,9 @@ long prepare_to_swait_event(struct swait_queue_head *q, struct swait_queue *wait
}
raw_spin_unlock_irqrestore(&q->lock, flags);
+ if (!ret && state & TASK_NORMAL)
+ sdt_wait(&q->dmap);
+
return ret;
}
EXPORT_SYMBOL(prepare_to_swait_event);
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* Re: [RFC 00/14] DEPT(DEPendency Tracker)
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (13 preceding siblings ...)
2022-01-25 2:36 ` [RFC 14/14] dept: Apply SDT to swait Byungchul Park
@ 2022-01-27 1:00 ` Byungchul Park
2022-01-27 1:08 ` Patches(Dept) for v5.17-rc1 Byungchul Park
15 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:00 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
+cc
linux-mm@kvack.org
akpm@linux-foundation.org
mhocko@kernel.org
minchan@kernel.org
hannes@cmpxchg.org
vdavydov.dev@gmail.com
sj@kernel.org
jglisse@redhat.com
dennis@kernel.org
cl@linux.com
penberg@kernel.org
rientjes@google.com
vbabka@suse.cz
ngupta@vflare.org
linux-block@vger.kernel.org
axboe@kernel.dk
paolo.valente@linaro.org
josef@toxicpanda.com
linux-fsdevel@vger.kernel.org
viro@zeniv.linux.org.uk
jack@suse.cz
jlayton@kernel.org
dan.j.williams@intel.com
hch@infradead.org
djwong@kernel.org
dri-devel@lists.freedesktop.org
airlied@linux.ie
rodrigosiqueiramelo@gmail.com
melissa.srw@gmail.com
hamohammed.sa@gmail.com
--->8---
From 68ee7ab996fc7d67b6b506f48da106493ca2546a Mon Sep 17 00:00:00 2001
From: Byungchul Park <byungchul.park@lge.com>
Date: Tue, 25 Jan 2022 10:12:54 +0900
Subject: [RFC 00/14] DEPT(DEPendency Tracker)
Hi forks,
I've been developing a tool for detecting deadlock possibilities by
tracking wait/event rather than lock(?) acquisition order to try to
cover all synchonization machanisms. It's done on v5.10 tag. I bet
it would work great! Try it and see what's gonna happen.
Now that there's a porting issue, I made Dept rely on Lockdep. But it
should be separated from Lockdep once it's considered worth having.
Benifit:
0. Works with all lock primitives.
1. Works with wait_for_completion()/complete().
2. Works with 'wait' on PG_locked.
3. Works with 'wait' on PG_writeback.
4. Works with swait/wakeup.
5. Multiple reports are allowed.
6. Deduplication control on multiple reports.
7. Withstand false positives thanks to 5.
8. Easy to tag any wait/event.
Future work:
0. To make it more stable.
1. To separates Dept from Lockdep.
2. To improves performance in terms of time and space.
3. To use Dept as a dependency engine for Lockdep.
4. To add any missing tags of wait/event in the kernel.
5. To deduplicate stack trace.
I hope you guys are gonna be satisfied with Dept. Don't hesitate to
give any feedback. I will adopt any feedbacks if reasonable.
Thanks,
Byungchul
Byungchul Park (14):
llist: Move llist_{head,node} definition to types.h
dept: Implement Dept(Dependency Tracker)
dept: Embed Dept data in Lockdep
dept: Apply Dept to spinlock
dept: Apply Dept to mutex families
dept: Apply Dept to rwlock
dept: Apply Dept to wait_for_completion()/complete()
dept: Apply Dept to seqlock
dept: Apply Dept to rwsem
dept: Add proc knobs to show stats and dependency graph
dept: Introduce split map concept and new APIs for them
dept: Apply Dept to wait/event of PG_{locked,writeback}
dept: Separate out SDT(Single-event Dependency Tracker) header
dept: Apply SDT to swait
include/linux/completion.h | 48 +-
include/linux/dept.h | 541 ++++++++
include/linux/dept_page.h | 71 +
include/linux/dept_sdt.h | 53 +
include/linux/hardirq.h | 3 +
include/linux/irqflags.h | 33 +-
include/linux/llist.h | 9 +-
include/linux/lockdep.h | 156 ++-
include/linux/lockdep_types.h | 3 +
include/linux/mutex.h | 31 +
include/linux/page-flags.h | 26 +-
include/linux/pagemap.h | 7 +-
include/linux/percpu-rwsem.h | 10 +-
include/linux/rtmutex.h | 11 +-
include/linux/rwlock.h | 48 +
include/linux/rwlock_api_smp.h | 8 +-
include/linux/rwlock_types.h | 7 +
include/linux/rwsem.h | 31 +
include/linux/sched.h | 3 +
include/linux/seqlock.h | 19 +-
include/linux/spinlock.h | 24 +
include/linux/spinlock_types.h | 10 +
include/linux/swait.h | 4 +
include/linux/types.h | 8 +
init/init_task.c | 2 +
init/main.c | 4 +
kernel/Makefile | 1 +
kernel/dependency/Makefile | 5 +
kernel/dependency/dept.c | 2593 +++++++++++++++++++++++++++++++++++++
kernel/dependency/dept_hash.h | 11 +
kernel/dependency/dept_internal.h | 26 +
kernel/dependency/dept_object.h | 14 +
kernel/dependency/dept_proc.c | 97 ++
kernel/exit.c | 1 +
kernel/fork.c | 2 +
kernel/locking/lockdep.c | 12 +-
kernel/module.c | 2 +
kernel/sched/completion.c | 12 +-
kernel/sched/swait.c | 8 +
kernel/softirq.c | 6 +-
kernel/trace/trace_preemptirq.c | 19 +-
lib/Kconfig.debug | 21 +
mm/filemap.c | 62 +
mm/page_ext.c | 5 +
44 files changed, 4009 insertions(+), 58 deletions(-)
create mode 100644 include/linux/dept.h
create mode 100644 include/linux/dept_page.h
create mode 100644 include/linux/dept_sdt.h
create mode 100644 kernel/dependency/Makefile
create mode 100644 kernel/dependency/dept.c
create mode 100644 kernel/dependency/dept_hash.h
create mode 100644 kernel/dependency/dept_internal.h
create mode 100644 kernel/dependency/dept_object.h
create mode 100644 kernel/dependency/dept_proc.c
--
1.9.1
^ permalink raw reply [flat|nested] 31+ messages in thread
* Patches(Dept) for v5.17-rc1
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
` (14 preceding siblings ...)
2022-01-27 1:00 ` [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
@ 2022-01-27 1:08 ` Byungchul Park
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
15 siblings, 1 reply; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:08 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
The following messages are for the latest tag, v5.17-rc1.
Hope you are gonna be satisfied with it!
Thanks,
Byungchul
--->8---
From 68ee7ab996fc7d67b6b506f48da106493ca2546a Mon Sep 17 00:00:00 2001
From: Byungchul Park <byungchul.park@lge.com>
Date: Tue, 25 Jan 2022 10:12:54 +0900
Subject: [RFC 00/14] DEPT(DEPendency Tracker)
Hi forks,
I've been developing a tool for detecting deadlock possibilities by
tracking wait/event rather than lock(?) acquisition order to try to
cover all synchonization machanisms. It's done on v5.10 tag. I bet
it would work great! Try it and see what's gonna happen.
Now that there's a porting issue, I made Dept rely on Lockdep. But it
should be separated from Lockdep once it's considered worth having.
Benifit:
0. Works with all lock primitives.
1. Works with wait_for_completion()/complete().
2. Works with 'wait' on PG_locked.
3. Works with 'wait' on PG_writeback.
4. Works with swait/wakeup.
5. Multiple reports are allowed.
6. Deduplication control on multiple reports.
7. Withstand false positives thanks to 5.
8. Easy to tag any wait/event.
Future work:
0. To make it more stable.
1. To separates Dept from Lockdep.
2. To improves performance in terms of time and space.
3. To use Dept as a dependency engine for Lockdep.
4. To add any missing tags of wait/event in the kernel.
5. To deduplicate stack trace.
I hope you guys are gonna be satisfied with Dept. Don't hesitate to
give any feedback. I will adopt any feedbacks if reasonable.
Thanks,
Byungchul
Byungchul Park (14):
llist: Move llist_{head,node} definition to types.h
dept: Implement Dept(Dependency Tracker)
dept: Embed Dept data in Lockdep
dept: Apply Dept to spinlock
dept: Apply Dept to mutex families
dept: Apply Dept to rwlock
dept: Apply Dept to wait_for_completion()/complete()
dept: Apply Dept to seqlock
dept: Apply Dept to rwsem
dept: Add proc knobs to show stats and dependency graph
dept: Introduce split map concept and new APIs for them
dept: Apply Dept to wait/event of PG_{locked,writeback}
dept: Separate out SDT(Single-event Dependency Tracker) header
dept: Apply SDT to swait
include/linux/completion.h | 48 +-
include/linux/dept.h | 541 ++++++++
include/linux/dept_page.h | 71 +
include/linux/dept_sdt.h | 53 +
include/linux/hardirq.h | 3 +
include/linux/irqflags.h | 33 +-
include/linux/llist.h | 9 +-
include/linux/lockdep.h | 156 ++-
include/linux/lockdep_types.h | 3 +
include/linux/mutex.h | 31 +
include/linux/page-flags.h | 26 +-
include/linux/pagemap.h | 7 +-
include/linux/percpu-rwsem.h | 10 +-
include/linux/rtmutex.h | 11 +-
include/linux/rwlock.h | 48 +
include/linux/rwlock_api_smp.h | 8 +-
include/linux/rwlock_types.h | 7 +
include/linux/rwsem.h | 31 +
include/linux/sched.h | 3 +
include/linux/seqlock.h | 19 +-
include/linux/spinlock.h | 24 +
include/linux/spinlock_types.h | 10 +
include/linux/swait.h | 4 +
include/linux/types.h | 8 +
init/init_task.c | 2 +
init/main.c | 4 +
kernel/Makefile | 1 +
kernel/dependency/Makefile | 5 +
kernel/dependency/dept.c | 2593 +++++++++++++++++++++++++++++++++++++
kernel/dependency/dept_hash.h | 11 +
kernel/dependency/dept_internal.h | 26 +
kernel/dependency/dept_object.h | 14 +
kernel/dependency/dept_proc.c | 97 ++
kernel/exit.c | 1 +
kernel/fork.c | 2 +
kernel/locking/lockdep.c | 12 +-
kernel/module.c | 2 +
kernel/sched/completion.c | 12 +-
kernel/sched/swait.c | 8 +
kernel/softirq.c | 6 +-
kernel/trace/trace_preemptirq.c | 19 +-
lib/Kconfig.debug | 21 +
mm/filemap.c | 62 +
mm/page_ext.c | 5 +
44 files changed, 4009 insertions(+), 58 deletions(-)
create mode 100644 include/linux/dept.h
create mode 100644 include/linux/dept_page.h
create mode 100644 include/linux/dept_sdt.h
create mode 100644 kernel/dependency/Makefile
create mode 100644 kernel/dependency/dept.c
create mode 100644 kernel/dependency/dept_hash.h
create mode 100644 kernel/dependency/dept_internal.h
create mode 100644 kernel/dependency/dept_object.h
create mode 100644 kernel/dependency/dept_proc.c
--
1.9.1
^ permalink raw reply [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h
2022-01-27 1:08 ` Patches(Dept) for v5.17-rc1 Byungchul Park
@ 2022-01-27 1:10 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 02/14] dept: Implement Dept(Dependency Tracker) Byungchul Park
` (12 more replies)
0 siblings, 13 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:10 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
llist_head and llist_node can be used by very primitives. For example,
Dept for tracking dependency uses llist things in its header. To avoid
header dependency, move those to types.h.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/llist.h | 8 --------
include/linux/types.h | 8 ++++++++
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/include/linux/llist.h b/include/linux/llist.h
index 85bda2d..99cc3c3 100644
--- a/include/linux/llist.h
+++ b/include/linux/llist.h
@@ -53,14 +53,6 @@
#include <linux/stddef.h>
#include <linux/types.h>
-struct llist_head {
- struct llist_node *first;
-};
-
-struct llist_node {
- struct llist_node *next;
-};
-
#define LLIST_HEAD_INIT(name) { NULL }
#define LLIST_HEAD(name) struct llist_head name = LLIST_HEAD_INIT(name)
diff --git a/include/linux/types.h b/include/linux/types.h
index ac825ad..4662d6e 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -187,6 +187,14 @@ struct hlist_node {
struct hlist_node *next, **pprev;
};
+struct llist_head {
+ struct llist_node *first;
+};
+
+struct llist_node {
+ struct llist_node *next;
+};
+
struct ustat {
__kernel_daddr_t f_tfree;
#ifdef CONFIG_ARCH_32BIT_USTAT_F_TINODE
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 02/14] dept: Implement Dept(Dependency Tracker)
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 03/14] dept: Embed Dept data in Lockdep Byungchul Park
` (11 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
CURRENT STATUS
--------------
Lockdep tracks acquisition order of locks in order to detect deadlock,
and IRQ and IRQ enable/disable state as well to take accident
acquisitions into account.
Lockdep should be turned off once it detects and reports a deadlock
since the data structure and algorithm are not reusable after detection
because of the complex design.
PROBLEM
-------
*Waits* and their *events* that never reach eventually cause deadlock.
However, Lockdep is only interested in lock acquisition order, forcing
to emulate lock acqusition even for just waits and events that have
nothing to do with real lock.
Even worse, no one likes Lockdep's false positive detection because that
prevents further one that might be more valuable. That's why all the
kernel developers are sensitive to Lockdep's false positive.
Besides those, by tracking acquisition order, it cannot correctly deal
with read lock and cross-event e.g. wait_for_completion()/complete() for
deadlock detection. Lockdep is no longer a good tool for that purpose.
SOLUTION
--------
Again, *waits* and their *events* that never reach eventually cause
deadlock. The new solution, Dept(DEPendency Tracker), focuses on waits
and events themselves. Dept tracks waits and events and report it if
any event would be never reachable.
Dept does:
. Works with read lock in the right way.
. Works with any wait and event e.i. cross-event.
. Continue to work even after reporting multiple times.
. Provides simple and intuitive APIs.
. Does exactly what dependency checker should do.
Q & A
-----
Q. Is this the first try ever to address the problem?
A. No. Cross-release feature (b09be676e0ff2 locking/lockdep: Implement
the 'crossrelease' feature) addressed it 2 years ago that was a
Lockdep extention and merged but reverted shortly because:
Cross-release started to report valuable hidden problems but started
to give report false positive reports as well. For sure, no one
likes Lockdep's false positive reports since it makes Lockdep stop,
preventing reporting further real problems.
Q. Why not Dept was developed as an extention of Lockdep?
A. Lockdep definitely includes all the efforts great developers have
made for a long time so as to be quite stable enough. But I had to
design and implement newly because of the following:
1) Lockdep was designed to track lock acquisition order. The APIs and
implementation do not fit on wait-event model.
2) Lockdep is turned off on detection including false positive. Which
is terrible and prevents developing any extention for stronger
detection.
Q. Do you intend to totally replace Lockdep?
A. No. Lockdep also checks if lock usage is correct. Of course, the
dependency check routine should be replaced but the other functions
should be still there.
Q. Do you mean the dependency check routine should be replaced right
away?
A. No. I admit Lockdep is stable enough thanks to great efforts kernel
developers have made. Lockdep and Dept, both should be in the kernel
until Dept gets considered stable.
Q. Stronger detection capability would give more false positive report.
Which was a big problem when cross-release was introduced. Is it ok
with Dept?
A. It's ok. Dept allows multiple reporting thanks to simple and quite
generalized design. Of course, false positive reports should be fixed
anyway but it's no longer as a critical problem as it was.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept.h | 541 +++++++++
include/linux/hardirq.h | 3 +
include/linux/irqflags.h | 33 +-
include/linux/sched.h | 3 +
init/init_task.c | 2 +
init/main.c | 2 +
kernel/Makefile | 1 +
kernel/dependency/Makefile | 4 +
kernel/dependency/dept.c | 2477 +++++++++++++++++++++++++++++++++++++++
kernel/dependency/dept_hash.h | 11 +
kernel/dependency/dept_object.h | 14 +
kernel/exit.c | 1 +
kernel/fork.c | 2 +
kernel/module.c | 2 +
kernel/softirq.c | 6 +-
kernel/trace/trace_preemptirq.c | 19 +-
lib/Kconfig.debug | 20 +
17 files changed, 3132 insertions(+), 9 deletions(-)
create mode 100644 include/linux/dept.h
create mode 100644 kernel/dependency/Makefile
create mode 100644 kernel/dependency/dept.c
create mode 100644 kernel/dependency/dept_hash.h
create mode 100644 kernel/dependency/dept_object.h
diff --git a/include/linux/dept.h b/include/linux/dept.h
new file mode 100644
index 0000000..8665629
--- /dev/null
+++ b/include/linux/dept.h
@@ -0,0 +1,541 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept(DEPendency Tracker) - runtime dependency tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ */
+
+#ifndef __LINUX_DEPT_H
+#define __LINUX_DEPT_H
+
+#ifdef CONFIG_DEPT
+
+#include <linux/types.h>
+
+struct task_struct;
+
+#define DEPT_MAX_STACK_ENTRY 16
+#define DEPT_MAX_WAIT_HIST 64
+#define DEPT_MAX_ECXT_HELD 48
+
+#define DEPT_MAX_SUBCLASSES 16
+#define DEPT_MAX_SUBCLASSES_EVT 2
+#define DEPT_MAX_SUBCLASSES_USR (DEPT_MAX_SUBCLASSES / DEPT_MAX_SUBCLASSES_EVT)
+#define DEPT_MAX_SUBCLASSES_CACHE 2
+
+#define DEPT_SIRQ 0
+#define DEPT_HIRQ 1
+#define DEPT_IRQS_NR 2
+#define DEPT_SIRQF (1UL << DEPT_SIRQ)
+#define DEPT_HIRQF (1UL << DEPT_HIRQ)
+
+struct dept_class {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * unique information about the class
+ */
+ const char *name;
+ unsigned long key;
+ int sub;
+
+ /*
+ * for BFS
+ */
+ unsigned int bfs_gen;
+ int bfs_dist;
+ struct dept_class *bfs_parent;
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking all classes
+ */
+ struct list_head all_node;
+
+ /*
+ * for associating its dependencies
+ */
+ struct list_head dep_head;
+ struct list_head dep_rev_head;
+
+ /*
+ * for tracking IRQ dependencies
+ */
+ int iwait_dist[DEPT_IRQS_NR];
+ struct dept_ecxt *iecxt[DEPT_IRQS_NR];
+ struct dept_wait *iwait[DEPT_IRQS_NR];
+};
+
+struct dept_stack {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * backtrace entries
+ */
+ unsigned long raw[DEPT_MAX_STACK_ENTRY];
+ int nr;
+};
+
+struct dept_ecxt {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * function that entered to this ecxt
+ */
+ const char *ecxt_fn;
+
+ /*
+ * event function
+ */
+ const char *event_fn;
+
+ /*
+ * associated class
+ */
+ struct dept_class *class;
+
+ /*
+ * flag indicating which IRQ has been
+ * enabled within the event context
+ */
+ unsigned long enirqf;
+
+ /*
+ * where the IRQ-enabled happened
+ */
+ unsigned long enirq_ip[DEPT_IRQS_NR];
+ struct dept_stack *enirq_stack[DEPT_IRQS_NR];
+
+ /*
+ * where the event context started
+ */
+ unsigned long ecxt_ip;
+ struct dept_stack *ecxt_stack;
+
+ /*
+ * where the event triggered
+ */
+ unsigned long event_ip;
+ struct dept_stack *event_stack;
+};
+
+struct dept_wait {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * function causing this wait
+ */
+ const char *wait_fn;
+
+ /*
+ * the associated class
+ */
+ struct dept_class *class;
+
+ /*
+ * which IRQ the wait was placed in
+ */
+ unsigned long irqf;
+
+ /*
+ * where the IRQ wait happened
+ */
+ unsigned long irq_ip[DEPT_IRQS_NR];
+ struct dept_stack *irq_stack[DEPT_IRQS_NR];
+
+ /*
+ * where the wait happened
+ */
+ unsigned long wait_ip;
+ struct dept_stack *wait_stack;
+};
+
+struct dept_staleiw {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking all staleiws
+ */
+ struct list_head all_node;
+
+ /*
+ * actual data
+ */
+ unsigned long ip;
+ int irq;
+};
+
+struct dept_staleie {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking all staleies
+ */
+ struct list_head all_node;
+
+ /*
+ * actual data
+ */
+ unsigned long ip;
+ int irq;
+};
+
+struct dept_dep {
+ union {
+ struct llist_node pool_node;
+
+ /*
+ * reference counter for object management
+ */
+ atomic_t ref;
+ };
+
+ /*
+ * key data of dependency
+ */
+ struct dept_ecxt *ecxt;
+ struct dept_wait *wait;
+
+ /*
+ * This object can be referred without dept_lock
+ * held but with IRQ disabled, e.g. for hash
+ * lookup. So deferred deletion is needed.
+ */
+ struct rcu_head rh;
+
+ /*
+ * for BFS
+ */
+ struct list_head bfs_node;
+
+ /*
+ * for hashing this object
+ */
+ struct hlist_node hash_node;
+
+ /*
+ * for linking to a class object
+ */
+ struct list_head dep_node;
+ struct list_head dep_rev_node;
+};
+
+struct dept_hash {
+ /*
+ * hash table
+ */
+ struct hlist_head *table;
+
+ /*
+ * size of the table e.i. 2^bits
+ */
+ int bits;
+};
+
+struct dept_pool {
+ const char *name;
+
+ /*
+ * object size
+ */
+ size_t obj_sz;
+
+ /*
+ * the number of the static array
+ */
+ atomic_t obj_nr;
+
+ /*
+ * offset of ->pool_node
+ */
+ size_t node_off;
+
+ /*
+ * pointer to the pool
+ */
+ void *spool;
+ struct llist_head boot_pool;
+ struct llist_head __percpu *lpool;
+};
+
+struct dept_ecxt_held {
+ /*
+ * associated event context
+ */
+ struct dept_ecxt *ecxt;
+
+ /*
+ * unique key for this dept_ecxt_held
+ */
+ unsigned long key;
+
+ /*
+ * the wgen when the event context started
+ */
+ unsigned int wgen;
+
+ /*
+ * for allowing user aware nesting
+ */
+ int nest;
+};
+
+struct dept_wait_hist {
+ /*
+ * associated wait
+ */
+ struct dept_wait *wait;
+
+ /*
+ * unique id of all waits system-wise until wrapped
+ */
+ unsigned int wgen;
+
+ /*
+ * local context id to identify IRQ context
+ */
+ unsigned int ctxt_id;
+};
+
+struct dept_task {
+ /*
+ * all event contexts that have entered and before exiting
+ */
+ struct dept_ecxt_held ecxt_held[DEPT_MAX_ECXT_HELD];
+ int ecxt_held_pos;
+
+ /*
+ * ring buffer holding all waits that have happened
+ */
+ struct dept_wait_hist wait_hist[DEPT_MAX_WAIT_HIST];
+ int wait_hist_pos;
+
+ /*
+ * sequencial id to identify each IRQ context
+ */
+ unsigned int irq_id[DEPT_IRQS_NR];
+
+ /*
+ * for tracking IRQ-enabled points with cross-event
+ */
+ unsigned int wgen_enirq[DEPT_IRQS_NR];
+
+ /*
+ * for keeping up-to-date IRQ-enabled points
+ */
+ unsigned long enirq_ip[DEPT_IRQS_NR];
+
+ /*
+ * current effective IRQ-enabled flag
+ */
+ unsigned long eff_enirqf;
+
+ /*
+ * for reserving a current stack instance at each operation
+ */
+ struct dept_stack *stack;
+
+ /*
+ * for preventing recursive call into Dept engine
+ */
+ int recursive;
+
+ /*
+ * for tracking IRQ-enable state
+ */
+ bool hardirqs_enabled;
+ bool softirqs_enabled;
+};
+
+struct dept_key {
+ union {
+ /*
+ * Each byte-wise address will be used as its key.
+ */
+ char subkeys[DEPT_MAX_SUBCLASSES];
+
+ /*
+ * for caching the main class pointer
+ */
+ struct dept_class *classes[DEPT_MAX_SUBCLASSES_CACHE];
+ };
+};
+
+struct dept_map {
+ const char *name;
+ struct dept_key *keys;
+ int sub_usr;
+
+ /*
+ * It's local copy for fast acces to the associated classes. And
+ * Also used for dept_key instance for statically defined map.
+ */
+ struct dept_key keys_local;
+
+ /*
+ * wait timestamp associated to this map
+ */
+ unsigned int wgen;
+
+ /*
+ * whether this map should be going to be checked or not
+ */
+ bool nocheck;
+};
+
+#define DEPT_TASK_INITIALIZER(t) \
+ .dept_task.wait_hist = { { .wait = NULL, } }, \
+ .dept_task.ecxt_held_pos = 0, \
+ .dept_task.wait_hist_pos = 0, \
+ .dept_task.irq_id = { 0 }, \
+ .dept_task.wgen_enirq = { 0 }, \
+ .dept_task.enirq_ip = { 0 }, \
+ .dept_task.recursive = 0, \
+ .dept_task.hardirqs_enabled = false, \
+ .dept_task.softirqs_enabled = false,
+
+extern void dept_on(void);
+extern void dept_off(void);
+extern void dept_init(void);
+extern void dept_task_init(struct task_struct *t);
+extern void dept_task_exit(struct task_struct *t);
+extern void dept_free_range(void *start, unsigned int sz);
+extern void dept_map_init(struct dept_map *m, struct dept_key *k, int sub, const char *n);
+extern void dept_map_reinit(struct dept_map *m);
+extern void dept_map_nocheck(struct dept_map *m);
+
+extern void dept_wait(struct dept_map *m, unsigned long w_f, unsigned long ip, const char *w_fn, int ne);
+extern void dept_ecxt_enter(struct dept_map *m, unsigned long e_f, unsigned long ip, const char *c_fn, const char *e_fn, int ne);
+extern void dept_asked_event(struct dept_map *m);
+extern void dept_event(struct dept_map *m, unsigned long e_f, unsigned long ip, const char *e_fn);
+extern void dept_ecxt_exit(struct dept_map *m, unsigned long ip);
+extern struct dept_map *dept_top_map(void);
+extern void dept_warn_on(bool cond);
+
+/*
+ * for users who want to manage external keys
+ */
+extern void dept_key_init(struct dept_key *k);
+extern void dept_key_destroy(struct dept_key *k);
+
+#define DEPT_SDT_MAP_INIT(dname) { .name = #dname }
+#define DEFINE_DEPT_SDT(x) \
+ struct dept_map x = DEPT_SDT_MAP_INIT(x)
+
+/*
+ * SDT(Single-event Dependency Tracker) APIs
+ *
+ * In case that one dept_map instance maps to a single event, SDT APIs
+ * can be used.
+ */
+#define sdt_map_init(m) \
+ do { \
+ static struct dept_key __key; \
+ dept_map_init(m, &__key, 0, #m); \
+ } while (0)
+#define sdt_map_init_key(m, k) dept_map_init(m, k, 0, #m)
+
+#define sdt_wait(m) \
+ do { \
+ dept_asked_event(m); \
+ dept_wait(m, 1UL, _THIS_IP_, "wait", 0); \
+ } while (0)
+#define sdt_ecxt_enter(m) dept_ecxt_enter(m, 1UL, _THIS_IP_, "start", "event", 0)
+#define sdt_event(m) dept_event(m, 1UL, _THIS_IP_, "event")
+#define sdt_ecxt_exit(m) dept_ecxt_exit(m, _THIS_IP_)
+#else /* !CONFIG_DEPT */
+struct dept_task { };
+struct dept_key { };
+struct dept_map { };
+
+#define DEPT_TASK_INITIALIZER(t)
+
+#define dept_on() do { } while (0)
+#define dept_off() do { } while (0)
+#define dept_init() do { } while (0)
+#define dept_task_init(t) do { } while (0)
+#define dept_task_exit(t) do { } while (0)
+#define dept_free_range(s, sz) do { } while (0)
+#define dept_map_init(m, k, s, n) do { (void)(n); (void)(k); } while (0)
+#define dept_map_reinit(m) do { } while (0)
+#define dept_map_nocheck(m) do { } while (0)
+
+#define dept_wait(m, w_f, ip, w_fn, ne) do { (void)(w_fn); } while (0)
+#define dept_ecxt_enter(m, e_f, ip, c_fn, e_fn, ne) do { (void)(c_fn); (void)(e_fn); } while (0)
+#define dept_asked_event(m) do { } while (0)
+#define dept_event(m, e_f, ip, e_fn) do { (void)(e_fn); } while (0)
+#define dept_ecxt_exit(m, ip) do { } while (0)
+#define dept_top_map() NULL
+#define dept_warn_on(c) do { } while (0)
+#define dept_key_init(k) do { (void)(k); } while (0)
+#define dept_key_destroy(k) do { (void)(k); } while (0)
+
+#define DEPT_SDT_MAP_INIT(dname)
+#define DEFINE_DEPT_SDT(x)
+
+#define sdt_map_init(m) do { } while (0)
+#define sdt_map_init_key(m, k) do { (void)(k); } while (0)
+#define sdt_wait(m) do { } while (0)
+#define sdt_ecxt_enter(m) do { } while (0)
+#define sdt_event(m) do { } while (0)
+#define sdt_ecxt_exit(m) do { } while (0)
+#endif
+#endif /* __LINUX_DEPT_H */
diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index 76878b3..07005f2 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
@@ -5,6 +5,7 @@
#include <linux/context_tracking_state.h>
#include <linux/preempt.h>
#include <linux/lockdep.h>
+#include <linux/dept.h>
#include <linux/ftrace_irq.h>
#include <linux/sched.h>
#include <linux/vtime.h>
@@ -114,6 +115,7 @@ static inline void rcu_nmi_exit(void) { }
*/
#define __nmi_enter() \
do { \
+ dept_off(); \
lockdep_off(); \
arch_nmi_enter(); \
BUG_ON(in_nmi() == NMI_MASK); \
@@ -136,6 +138,7 @@ static inline void rcu_nmi_exit(void) { }
__preempt_count_sub(NMI_OFFSET + HARDIRQ_OFFSET); \
arch_nmi_exit(); \
lockdep_on(); \
+ dept_on(); \
} while (0)
#define nmi_exit() \
diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index 4b14093..c5c7f47 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -31,6 +31,22 @@
static inline void lockdep_hardirqs_off(unsigned long ip) { }
#endif
+#ifdef CONFIG_DEPT
+ extern void dept_hardirq_enter(void);
+ extern void dept_softirq_enter(void);
+ extern void dept_enable_hardirq(unsigned long ip);
+ extern void dept_enable_softirq(unsigned long ip);
+ extern void dept_disable_hardirq(unsigned long ip);
+ extern void dept_disable_softirq(unsigned long ip);
+#else
+ static inline void dept_hardirq_enter(void) { }
+ static inline void dept_softirq_enter(void) { }
+ static inline void dept_enable_hardirq(unsigned long ip) { }
+ static inline void dept_enable_softirq(unsigned long ip) { }
+ static inline void dept_disable_hardirq(unsigned long ip) { }
+ static inline void dept_disable_softirq(unsigned long ip) { }
+#endif
+
#ifdef CONFIG_TRACE_IRQFLAGS
/* Per-task IRQ trace events information. */
@@ -53,15 +69,19 @@ struct irqtrace_events {
extern void trace_hardirqs_off_finish(void);
extern void trace_hardirqs_on(void);
extern void trace_hardirqs_off(void);
+extern void trace_softirqs_on_caller(unsigned long ip);
+extern void trace_softirqs_off_caller(unsigned long ip);
# define lockdep_hardirq_context() (raw_cpu_read(hardirq_context))
# define lockdep_softirq_context(p) ((p)->softirq_context)
# define lockdep_hardirqs_enabled() (this_cpu_read(hardirqs_enabled))
# define lockdep_softirqs_enabled(p) ((p)->softirqs_enabled)
-# define lockdep_hardirq_enter() \
-do { \
- if (__this_cpu_inc_return(hardirq_context) == 1)\
- current->hardirq_threaded = 0; \
+# define lockdep_hardirq_enter() \
+do { \
+ if (__this_cpu_inc_return(hardirq_context) == 1) { \
+ current->hardirq_threaded = 0; \
+ dept_hardirq_enter(); \
+ } \
} while (0)
# define lockdep_hardirq_threaded() \
do { \
@@ -115,6 +135,8 @@ struct irqtrace_events {
# define trace_hardirqs_off_finish() do { } while (0)
# define trace_hardirqs_on() do { } while (0)
# define trace_hardirqs_off() do { } while (0)
+# define trace_softirqs_on_caller(ip) do { } while (0)
+# define trace_softirqs_off_caller(ip) do { } while (0)
# define lockdep_hardirq_context() 0
# define lockdep_softirq_context(p) 0
# define lockdep_hardirqs_enabled() 0
@@ -135,7 +157,8 @@ struct irqtrace_events {
#if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PREEMPT_RT)
# define lockdep_softirq_enter() \
do { \
- current->softirq_context++; \
+ if (!current->softirq_context++) \
+ dept_softirq_enter(); \
} while (0)
# define lockdep_softirq_exit() \
do { \
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 508b91d..0bcb342 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -35,6 +35,7 @@
#include <linux/seqlock.h>
#include <linux/kcsan.h>
#include <asm/kmap_size.h>
+#include <linux/dept.h>
/* task_struct member predeclarations (sorted alphabetically): */
struct audit_context;
@@ -1157,6 +1158,8 @@ struct task_struct {
struct held_lock held_locks[MAX_LOCK_DEPTH];
#endif
+ struct dept_task dept_task;
+
#if defined(CONFIG_UBSAN) && !defined(CONFIG_UBSAN_TRAP)
unsigned int in_ubsan;
#endif
diff --git a/init/init_task.c b/init/init_task.c
index 73cc8f0..d530256 100644
--- a/init/init_task.c
+++ b/init/init_task.c
@@ -12,6 +12,7 @@
#include <linux/audit.h>
#include <linux/numa.h>
#include <linux/scs.h>
+#include <linux/dept.h>
#include <linux/uaccess.h>
@@ -193,6 +194,7 @@ struct task_struct init_task
.curr_chain_key = INITIAL_CHAIN_KEY,
.lockdep_recursion = 0,
#endif
+ DEPT_TASK_INITIALIZER(init_task)
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
.ret_stack = NULL,
.tracing_graph_pause = ATOMIC_INIT(0),
diff --git a/init/main.c b/init/main.c
index 65fa2e4..ca96e11 100644
--- a/init/main.c
+++ b/init/main.c
@@ -65,6 +65,7 @@
#include <linux/debug_locks.h>
#include <linux/debugobjects.h>
#include <linux/lockdep.h>
+#include <linux/dept.h>
#include <linux/kmemleak.h>
#include <linux/padata.h>
#include <linux/pid_namespace.h>
@@ -1070,6 +1071,7 @@ asmlinkage __visible void __init __no_sanitize_address start_kernel(void)
panic_param);
lockdep_init();
+ dept_init();
/*
* Need to run this when irqs are enabled, because it wants
diff --git a/kernel/Makefile b/kernel/Makefile
index 56f4ee9..cef9b02 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -53,6 +53,7 @@ obj-y += rcu/
obj-y += livepatch/
obj-y += dma/
obj-y += entry/
+obj-y += dependency/
obj-$(CONFIG_KCMP) += kcmp.o
obj-$(CONFIG_FREEZER) += freezer.o
diff --git a/kernel/dependency/Makefile b/kernel/dependency/Makefile
new file mode 100644
index 0000000..9f7778e
--- /dev/null
+++ b/kernel/dependency/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_DEPT) += dept.o
+
diff --git a/kernel/dependency/dept.c b/kernel/dependency/dept.c
new file mode 100644
index 0000000..64287894
--- /dev/null
+++ b/kernel/dependency/dept.c
@@ -0,0 +1,2477 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept(DEPendency Tracker) - Runtime dependency tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ *
+ * Dept provides a general way to detect deadlock possibility in runtime
+ * and the interest is not limited to typical lock but to every
+ * syncronization primitives.
+ *
+ * The following ideas were borrowed from Lockdep:
+ *
+ * 1) Use a graph to track relationship between classes.
+ * 2) Prevent performance regression using hash.
+ *
+ * The following items were enhanced from Lockdep:
+ *
+ * 1) Cover more deadlock cases.
+ * 2) Allow muliple reports.
+ *
+ * TODO: Both Lockdep and Dept should co-exist until Dept is considered
+ * stable. Then the dependency check routine shoud be replaced with Dept
+ * after. It should finally look like:
+ *
+ *
+ *
+ * As is:
+ *
+ * Lockdep
+ * +-----------------------------------------+
+ * | Lock usage correctness check | <-> locks
+ * | |
+ * | |
+ * | +-------------------------------------+ |
+ * | | Dependency check | |
+ * | | (by tracking lock aquisition order) | |
+ * | +-------------------------------------+ |
+ * | |
+ * +-----------------------------------------+
+ *
+ * Dept
+ * +-----------------------------------------+
+ * | Dependency check | <-> waits/events
+ * | (by tracking wait and event context) |
+ * +-----------------------------------------+
+ *
+ *
+ *
+ * To be:
+ *
+ * Lockdep
+ * +-----------------------------------------+
+ * | Lock usage correctness check | <-> locks
+ * | |
+ * | |
+ * | (Request dependency check) |
+ * | T |
+ * +--------------------|--------------------+
+ * |
+ * Dept V
+ * +-----------------------------------------+
+ * | Dependency check | <-> waits/events
+ * | (by tracking wait and event context) |
+ * +-----------------------------------------+
+ *
+ *
+ *
+ * ---
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your ootion) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, you can access it online at
+ * http://www.gnu.org/licenses/gpl-2.0.html.
+ */
+
+#include <linux/sched.h>
+#include <linux/stacktrace.h>
+#include <linux/spinlock.h>
+#include <linux/kallsyms.h>
+#include <linux/hash.h>
+#include <linux/dept.h>
+#include <linux/utsname.h>
+
+static int dept_stop;
+static int dept_per_cpu_ready;
+
+/*
+ * Make all operations using DEPT_WARN_ON() fail on oops_in_progress and
+ * prevent warning message.
+ */
+#define DEPT_WARN_ON_ONCE(c) \
+ ({ \
+ int __ret = 1; \
+ if (likely(!oops_in_progress)) \
+ __ret = WARN_ONCE(c, "DEPT_WARN_ON_ONCE: " #c); \
+ __ret; \
+ })
+
+#define DEPT_WARN_ONCE(s...) \
+ ({ \
+ if (likely(!oops_in_progress)) \
+ WARN_ONCE(1, "DEPT_WARN_ONCE: " s); \
+ })
+
+#define DEPT_WARN_ON(c) \
+ ({ \
+ int __ret = 1; \
+ if (likely(!oops_in_progress)) \
+ __ret = WARN(c, "DEPT_WARN_ON: " #c); \
+ __ret; \
+ })
+
+#define DEPT_WARN(s...) \
+ ({ \
+ if (likely(!oops_in_progress)) \
+ WARN(1, "DEPT_WARN: " s); \
+ })
+
+#define DEPT_STOP(s...) \
+ ({ \
+ WRITE_ONCE(dept_stop, 1); \
+ if (likely(!oops_in_progress)) \
+ WARN(1, "DEPT_STOP: " s); \
+ })
+
+static arch_spinlock_t dept_spin = (arch_spinlock_t)__ARCH_SPIN_LOCK_UNLOCKED;
+
+/*
+ * Dept internal engine should be careful in using outside functions
+ * e.g. printk at reporting since that kind of usage might cause
+ * untrackable deadlock.
+ */
+static atomic_t dept_outworld = ATOMIC_INIT(0);
+
+static inline void dept_outworld_enter(void)
+{
+ atomic_inc(&dept_outworld);
+}
+
+static inline void dept_outworld_exit(void)
+{
+ atomic_dec(&dept_outworld);
+}
+
+static inline bool dept_outworld_entered(void)
+{
+ return atomic_read(&dept_outworld);
+}
+
+static inline bool dept_lock(void)
+{
+ while (!arch_spin_trylock(&dept_spin))
+ if (unlikely(dept_outworld_entered()))
+ return false;
+ return true;
+}
+
+static inline void dept_unlock(void)
+{
+ arch_spin_unlock(&dept_spin);
+}
+
+/*
+ * whether to stack-trace on every wait or every ecxt
+ */
+static bool rich_stack = true;
+
+enum bfs_ret {
+ DEPT_BFS_CONTINUE,
+ DEPT_BFS_CONTINUE_REV,
+ DEPT_BFS_DONE,
+ DEPT_BFS_SKIP,
+};
+
+/*
+ * The irq wait is in unknown state. Should identify the state.
+ */
+#define DEPT_IWAIT_UNKNOWN ((void *)1)
+
+static inline bool before(unsigned int a, unsigned int b)
+{
+ return (int)(a - b) < 0;
+}
+
+static inline bool valid_stack(struct dept_stack *s)
+{
+ return s && s->nr > 0;
+}
+
+static inline bool valid_class(struct dept_class *c)
+{
+ return c->key;
+}
+
+static inline void invalidate_class(struct dept_class *c)
+{
+ c->key = 0UL;
+}
+
+static inline struct dept_class *dep_fc(struct dept_dep *d)
+{
+ return d->ecxt->class;
+}
+
+static inline struct dept_class *dep_tc(struct dept_dep *d)
+{
+ return d->wait->class;
+}
+
+static inline const char *irq_str(int irq)
+{
+ if (irq == DEPT_SIRQ)
+ return "softirq";
+ if (irq == DEPT_HIRQ)
+ return "hardirq";
+ return "(unknown)";
+}
+
+static inline struct dept_task *dept_task(void)
+{
+ return ¤t->dept_task;
+}
+
+/*
+ * Pool
+ * =====================================================================
+ * Dept maintains pools to provide objects in a safe way.
+ *
+ * 1) Static pool is used at the beginning of booting time.
+ * 2) Local pool is tried first before the static pool. Objects that
+ * have been freed will be placed.
+ */
+
+enum object_t {
+#define OBJECT(id, nr) OBJECT_##id,
+ #include "dept_object.h"
+#undef OBJECT
+ OBJECT_NR,
+};
+
+#define OBJECT(id, nr) \
+static struct dept_##id spool_##id[nr]; \
+static DEFINE_PER_CPU(struct llist_head, lpool_##id);
+ #include "dept_object.h"
+#undef OBJECT
+
+static struct dept_pool pool[OBJECT_NR] = {
+#define OBJECT(id, nr) { \
+ .name = #id, \
+ .obj_sz = sizeof(struct dept_##id), \
+ .obj_nr = ATOMIC_INIT(nr), \
+ .node_off = offsetof(struct dept_##id, pool_node), \
+ .spool = spool_##id, \
+ .lpool = &lpool_##id, },
+ #include "dept_object.h"
+#undef OBJECT
+};
+
+/*
+ * Can use llist no matter whether CONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG is
+ * enabled because Dept never race with NMI by nesting control.
+ */
+static void *from_pool(enum object_t t)
+{
+ struct dept_pool *p;
+ struct llist_head *h;
+ struct llist_node *n;
+
+ /*
+ * llist_del_first() doesn't allow concurrent access e.g.
+ * between process and IRQ context.
+ */
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return NULL;
+
+ p = &pool[t];
+
+ /*
+ * Try local pool first.
+ */
+ if (likely(dept_per_cpu_ready))
+ h = this_cpu_ptr(p->lpool);
+ else
+ h = &p->boot_pool;
+
+ n = llist_del_first(h);
+ if (n)
+ return (void *)n - p->node_off;
+
+ /*
+ * Try static pool.
+ */
+ if (atomic_read(&p->obj_nr) > 0) {
+ int idx = atomic_dec_return(&p->obj_nr);
+ if (idx >= 0)
+ return p->spool + (idx * p->obj_sz);
+ }
+
+ DEPT_WARN_ONCE("Pool(%s) is empty.\n", p->name);
+ return NULL;
+}
+
+static void to_pool(void *o, enum object_t t)
+{
+ struct dept_pool *p = &pool[t];
+ struct llist_head *h;
+
+ preempt_disable();
+ if (likely(dept_per_cpu_ready))
+ h = this_cpu_ptr(p->lpool);
+ else
+ h = &p->boot_pool;
+
+ llist_add(o + p->node_off, h);
+ preempt_enable();
+}
+
+#define OBJECT(id, nr) \
+static void (*ctor_##id)(struct dept_##id *a); \
+static void (*dtor_##id)(struct dept_##id *a); \
+static inline struct dept_##id *new_##id(void) \
+{ \
+ struct dept_##id *a; \
+ \
+ a = (struct dept_##id *)from_pool(OBJECT_##id); \
+ if (unlikely(!a)) \
+ return NULL; \
+ \
+ atomic_set(&a->ref, 1); \
+ \
+ if (ctor_##id) \
+ ctor_##id(a); \
+ \
+ return a; \
+} \
+ \
+static inline struct dept_##id *get_##id(struct dept_##id *a) \
+{ \
+ atomic_inc(&a->ref); \
+ return a; \
+} \
+ \
+static inline void put_##id(struct dept_##id *a) \
+{ \
+ if (!atomic_dec_return(&a->ref)) { \
+ if (dtor_##id) \
+ dtor_##id(a); \
+ to_pool(a, OBJECT_##id); \
+ } \
+} \
+ \
+static inline void del_##id(struct dept_##id *a) \
+{ \
+ put_##id(a); \
+} \
+ \
+static inline bool id##_refered(struct dept_##id *a, int expect) \
+{ \
+ return a && atomic_read(&a->ref) > expect; \
+}
+#include "dept_object.h"
+#undef OBJECT
+
+#define SET_CONSTRUCTOR(id, f) \
+static void (*ctor_##id)(struct dept_##id *a) = f;
+
+static void initialize_dep(struct dept_dep *d)
+{
+ INIT_LIST_HEAD(&d->bfs_node);
+ INIT_LIST_HEAD(&d->dep_node);
+ INIT_LIST_HEAD(&d->dep_rev_node);
+}
+SET_CONSTRUCTOR(dep, initialize_dep);
+
+static void initialize_class(struct dept_class *c)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ c->iwait_dist[i] = INT_MAX;
+ WRITE_ONCE(c->iecxt[i], NULL);
+ WRITE_ONCE(c->iwait[i], NULL);
+ }
+ c->bfs_gen = 0U;
+
+ INIT_LIST_HEAD(&c->all_node);
+ INIT_LIST_HEAD(&c->dep_head);
+ INIT_LIST_HEAD(&c->dep_rev_head);
+}
+SET_CONSTRUCTOR(class, initialize_class);
+
+static void initialize_ecxt(struct dept_ecxt *e)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ e->enirq_stack[i] = NULL;
+ e->enirq_ip[i] = 0UL;
+ }
+ e->ecxt_ip = 0UL;
+ e->ecxt_stack = NULL;
+ e->enirqf = 0UL;
+ e->event_stack = NULL;
+}
+SET_CONSTRUCTOR(ecxt, initialize_ecxt);
+
+static void initialize_wait(struct dept_wait *w)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ w->irq_stack[i] = NULL;
+ w->irq_ip[i] = 0UL;
+ }
+ w->wait_ip = 0UL;
+ w->wait_stack = NULL;
+ w->irqf = 0UL;
+}
+SET_CONSTRUCTOR(wait, initialize_wait);
+
+static void initialize_stack(struct dept_stack *s)
+{
+ s->nr = 0;
+}
+SET_CONSTRUCTOR(stack, initialize_stack);
+
+#define OBJECT(id, nr) \
+static void (*ctor_##id)(struct dept_##id *a);
+ #include "dept_object.h"
+#undef OBJECT
+
+#undef SET_CONSTRUCTOR
+
+#define SET_DESTRUCTOR(id, f) \
+static void (*dtor_##id)(struct dept_##id *a) = f;
+
+static void destroy_dep(struct dept_dep *d)
+{
+ if (d->ecxt)
+ put_ecxt(d->ecxt);
+ if (d->wait)
+ put_wait(d->wait);
+}
+SET_DESTRUCTOR(dep, destroy_dep);
+
+static void destroy_ecxt(struct dept_ecxt *e)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++)
+ if (e->enirq_stack[i])
+ put_stack(e->enirq_stack[i]);
+ if (e->class)
+ put_class(e->class);
+ if (e->ecxt_stack)
+ put_stack(e->ecxt_stack);
+ if (e->event_stack)
+ put_stack(e->event_stack);
+}
+SET_DESTRUCTOR(ecxt, destroy_ecxt);
+
+static void destroy_wait(struct dept_wait *w)
+{
+ int i;
+
+ for (i = 0; i < DEPT_IRQS_NR; i++)
+ if (w->irq_stack[i])
+ put_stack(w->irq_stack[i]);
+ if (w->class)
+ put_class(w->class);
+ if (w->wait_stack)
+ put_stack(w->wait_stack);
+}
+SET_DESTRUCTOR(wait, destroy_wait);
+
+#define OBJECT(id, nr) \
+static void (*dtor_##id)(struct dept_##id *a);
+ #include "dept_object.h"
+#undef OBJECT
+
+#undef SET_DESTRUCTOR
+
+/*
+ * Caching and hashing
+ * =====================================================================
+ * Dept makes use of caching and hashing to improve performance. Each
+ * object can be obtained in O(1) with its key.
+ *
+ * NOTE: Currently we assume all the objects in the hashs will never be
+ * removed. Implement it when needed.
+ */
+
+/*
+ * Some information might be lost but it's only for hashing key.
+ */
+static inline unsigned long mix(unsigned long a, unsigned long b)
+{
+ int halfbits = sizeof(unsigned long) * 8 / 2;
+ unsigned long halfmask = (1UL << halfbits) - 1UL;
+ return (a << halfbits) | (b & halfmask);
+}
+
+static bool cmp_dep(struct dept_dep *d1, struct dept_dep *d2)
+{
+ return dep_fc(d1)->key == dep_fc(d2)->key &&
+ dep_tc(d1)->key == dep_tc(d2)->key;
+}
+
+static unsigned long key_dep(struct dept_dep *d)
+{
+ return mix(dep_fc(d)->key, dep_tc(d)->key);
+}
+
+static bool cmp_class(struct dept_class *c1, struct dept_class *c2)
+{
+ return c1->key == c2->key;
+}
+
+static unsigned long key_class(struct dept_class *c)
+{
+ return c->key;
+}
+
+static bool cmp_staleiw(struct dept_staleiw *s1, struct dept_staleiw *s2)
+{
+ return s1->ip == s2->ip && s1->irq == s2->irq;
+}
+
+static unsigned long key_staleiw(struct dept_staleiw *s)
+{
+ return s->ip ^ (1UL << s->irq);
+}
+
+static bool cmp_staleie(struct dept_staleie *s1, struct dept_staleie *s2)
+{
+ return s1->ip == s2->ip && s1->irq == s2->irq;
+}
+
+static unsigned long key_staleie(struct dept_staleie *s)
+{
+ return s->ip ^ (1UL << s->irq);
+}
+
+#define HASH(id, bits) \
+static struct hlist_head table_##id[1UL << bits]; \
+ \
+static inline struct hlist_head *head_##id(struct dept_##id *a) \
+{ \
+ return table_##id + hash_long(key_##id(a), bits); \
+} \
+ \
+static inline struct dept_##id *hash_lookup_##id(struct dept_##id *a) \
+{ \
+ struct dept_##id *b; \
+ hlist_for_each_entry_rcu(b, head_##id(a), hash_node) \
+ if (cmp_##id(a, b)) \
+ return b; \
+ return NULL; \
+} \
+ \
+static inline void hash_add_##id(struct dept_##id *a) \
+{ \
+ hlist_add_head_rcu(&a->hash_node, head_##id(a)); \
+} \
+ \
+static inline void hash_del_##id(struct dept_##id *a) \
+{ \
+ hlist_del_rcu(&a->hash_node); \
+}
+#include "dept_hash.h"
+#undef HASH
+
+static inline struct dept_dep *dep(struct dept_class *fc,
+ struct dept_class *tc)
+{
+ struct dept_ecxt onetime_e = { .class = fc };
+ struct dept_wait onetime_w = { .class = tc };
+ struct dept_dep onetime_d = { .ecxt = &onetime_e,
+ .wait = &onetime_w };
+ return hash_lookup_dep(&onetime_d);
+}
+
+static inline struct dept_class *class(unsigned long key)
+{
+ struct dept_class onetime_c = { .key = key };
+ return hash_lookup_class(&onetime_c);
+}
+
+static inline struct dept_staleiw *staleiw(unsigned long ip, int irq)
+{
+ struct dept_staleiw onetime_s = { .ip = ip, .irq = irq };
+ return hash_lookup_staleiw(&onetime_s);
+}
+
+static inline struct dept_staleie *staleie(unsigned long ip, int irq)
+{
+ struct dept_staleie onetime_s = { .ip = ip, .irq = irq };
+ return hash_lookup_staleie(&onetime_s);
+}
+
+/*
+ * Report
+ * =====================================================================
+ * Dept prints useful information to help debuging on detection of
+ * problematic dependency.
+ */
+
+static inline void print_ip_stack(unsigned long ip, struct dept_stack *s)
+{
+ if (ip)
+ print_ip_sym(KERN_WARNING, ip);
+
+ if (valid_stack(s)) {
+ printk("stacktrace:\n");
+ stack_trace_print(s->raw, s->nr, 5);
+ }
+
+ if (!ip && !valid_stack(s))
+ printk("(N/A)\n");
+}
+
+#define print_spc(spc, fmt, ...) \
+ printk("%*c" fmt, (spc) * 4, ' ', ##__VA_ARGS__)
+
+static void print_diagram(struct dept_dep *d)
+{
+ struct dept_ecxt *e = d->ecxt;
+ struct dept_wait *w = d->wait;
+ struct dept_class *fc = dep_fc(d);
+ struct dept_class *tc = dep_tc(d);
+ unsigned long irqf;
+ int irq;
+ bool firstline = true;
+ int spc = 1;
+ const char *w_fn = w->wait_fn ?: "(unknown)";
+ const char *e_fn = e->event_fn ?: "(unknown)";
+ const char *c_fn = e->ecxt_fn ?: "(unknown)";
+
+ irqf = e->enirqf & w->irqf;
+
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR) {
+ if (!firstline)
+ printk("\nor\n\n");
+ firstline = false;
+
+ print_spc(spc, "[S] %s(%s:%d)\n", c_fn, fc->name, fc->sub);
+ print_spc(spc, " <%s interrupt>\n", irq_str(irq));
+ print_spc(spc + 1, "[W] %s(%s:%d)\n", w_fn, tc->name, tc->sub);
+ print_spc(spc, "[E] %s(%s:%d)\n", e_fn, fc->name, fc->sub);
+ }
+
+ if (!irqf) {
+ print_spc(spc, "[S] %s(%s:%d)\n", c_fn, fc->name, fc->sub);
+ print_spc(spc, "[W] %s(%s:%d)\n", w_fn, tc->name, tc->sub);
+ print_spc(spc, "[E] %s(%s:%d)\n", e_fn, fc->name, fc->sub);
+ }
+}
+
+static void print_dep(struct dept_dep *d)
+{
+ struct dept_ecxt *e = d->ecxt;
+ struct dept_wait *w = d->wait;
+ struct dept_class *fc = dep_fc(d);
+ struct dept_class *tc = dep_tc(d);
+ unsigned long irqf;
+ int irq;
+ const char *w_fn = w->wait_fn ?: "(unknown)";
+ const char *e_fn = e->event_fn ?: "(unknown)";
+ const char *c_fn = e->ecxt_fn ?: "(unknown)";
+
+ irqf = e->enirqf & w->irqf;
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR) {
+ printk("%s has been enabled:\n", irq_str(irq));
+ print_ip_stack(e->enirq_ip[irq], e->enirq_stack[irq]);
+ printk("\n");
+
+ printk("[S] %s(%s:%d):\n", c_fn, fc->name, fc->sub);
+ print_ip_stack(e->ecxt_ip, e->ecxt_stack);
+ printk("\n");
+
+ printk("[W] %s(%s:%d) in %s context:\n",
+ w_fn, tc->name, tc->sub, irq_str(irq));
+ print_ip_stack(w->irq_ip[irq], w->irq_stack[irq]);
+ printk("\n");
+
+ printk("[E] %s(%s:%d):\n", e_fn, fc->name, fc->sub);
+ print_ip_stack(e->event_ip, e->event_stack);
+ }
+
+ if (!irqf) {
+ printk("[S] %s(%s:%d):\n", c_fn, fc->name, fc->sub);
+ print_ip_stack(e->ecxt_ip, e->ecxt_stack);
+ printk("\n");
+
+ printk("[W] %s(%s:%d):\n", w_fn, tc->name, tc->sub);
+ print_ip_stack(w->wait_ip, w->wait_stack);
+ printk("\n");
+
+ printk("[E] %s(%s:%d):\n", e_fn, fc->name, fc->sub);
+ print_ip_stack(e->event_ip, e->event_stack);
+ }
+}
+
+static void save_current_stack(int skip);
+
+/*
+ * Print all classes in a circle.
+ */
+static void print_circle(struct dept_class *c)
+{
+ struct dept_class *fc = c->bfs_parent;
+ struct dept_class *tc = c;
+ int i;
+
+ dept_outworld_enter();
+ save_current_stack(6);
+
+ printk("===================================================\n");
+ printk("Dept: Circular dependency has been detected.\n");
+ printk("%s %.*s %s\n", init_utsname()->release,
+ (int)strcspn(init_utsname()->version, " "),
+ init_utsname()->version,
+ print_tainted());
+ printk("---------------------------------------------------\n");
+ printk("summary\n");
+ printk("---------------------------------------------------\n");
+
+ if (fc == tc)
+ printk("*** AA DEADLOCK ***\n\n");
+ else
+ printk("*** DEADLOCK ***\n\n");
+
+ i = 0;
+ do {
+ struct dept_dep *d = dep(fc, tc);
+
+ printk("context %c\n", 'A' + (i++));
+ print_diagram(d);
+ if (fc != c)
+ printk("\n");
+
+ tc = fc;
+ fc = fc->bfs_parent;
+ } while (tc != c);
+
+ printk("\n");
+ printk("[S]: start of the event context\n");
+ printk("[W]: the wait blocked\n");
+ printk("[E]: the event not reachable\n");
+
+ i = 0;
+ do {
+ struct dept_dep *d = dep(fc, tc);
+
+ printk("---------------------------------------------------\n");
+ printk("context %c's detail\n", 'A' + i);
+ printk("---------------------------------------------------\n");
+ printk("context %c\n", 'A' + (i++));
+ print_diagram(d);
+ printk("\n");
+ print_dep(d);
+
+ tc = fc;
+ fc = fc->bfs_parent;
+ } while (tc != c);
+
+ printk("---------------------------------------------------\n");
+ printk("information that might be helpful\n");
+ printk("---------------------------------------------------\n");
+ dump_stack();
+
+ dept_outworld_exit();
+}
+
+/*
+ * BFS(Breadth First Search)
+ * =====================================================================
+ * Whenever a new dependency is added into the graph, search the graph
+ * for a new circular dependency.
+ */
+
+static inline void enqueue(struct list_head *h, struct dept_dep *d)
+{
+ list_add_tail(&d->bfs_node, h);
+}
+
+static inline struct dept_dep *dequeue(struct list_head *h)
+{
+ struct dept_dep *d;
+ d = list_first_entry(h, struct dept_dep, bfs_node);
+ list_del(&d->bfs_node);
+ return d;
+}
+
+static inline bool empty(struct list_head *h)
+{
+ return list_empty(h);
+}
+
+static void extend_queue(struct list_head *h, struct dept_class *cur)
+{
+ struct dept_dep *d;
+
+ list_for_each_entry(d, &cur->dep_head, dep_node) {
+ struct dept_class *next = dep_tc(d);
+ if (cur->bfs_gen == next->bfs_gen)
+ continue;
+ next->bfs_gen = cur->bfs_gen;
+ next->bfs_dist = cur->bfs_dist + 1;
+ next->bfs_parent = cur;
+ enqueue(h, d);
+ }
+}
+
+static void extend_queue_rev(struct list_head *h, struct dept_class *cur)
+{
+ struct dept_dep *d;
+
+ list_for_each_entry(d, &cur->dep_rev_head, dep_rev_node) {
+ struct dept_class *next = dep_fc(d);
+ if (cur->bfs_gen == next->bfs_gen)
+ continue;
+ next->bfs_gen = cur->bfs_gen;
+ next->bfs_dist = cur->bfs_dist + 1;
+ next->bfs_parent = cur;
+ enqueue(h, d);
+ }
+}
+
+typedef enum bfs_ret bfs_f(struct dept_dep *d, void *in, void **out);
+static unsigned int bfs_gen;
+
+/*
+ * NOTE: Must be called with dept_lock held.
+ */
+static void bfs(struct dept_class *c, bfs_f *cb, void *in, void **out)
+{
+ LIST_HEAD(q);
+ enum bfs_ret ret;
+
+ if (DEPT_WARN_ON(!cb))
+ return;
+
+ /*
+ * Avoid zero bfs_gen.
+ */
+ bfs_gen = bfs_gen + 1 ?: 1;
+
+ c->bfs_gen = bfs_gen;
+ c->bfs_dist = 0;
+ c->bfs_parent = c;
+
+ ret = cb(NULL, in, out);
+ if (ret == DEPT_BFS_DONE)
+ return;
+ if (ret == DEPT_BFS_SKIP)
+ return;
+ if (ret == DEPT_BFS_CONTINUE)
+ extend_queue(&q, c);
+ if (ret == DEPT_BFS_CONTINUE_REV)
+ extend_queue_rev(&q, c);
+
+ while (!empty(&q)) {
+ struct dept_dep *d = dequeue(&q);
+
+ ret = cb(d, in, out);
+ if (ret == DEPT_BFS_DONE)
+ break;
+ if (ret == DEPT_BFS_SKIP)
+ continue;
+ if (ret == DEPT_BFS_CONTINUE)
+ extend_queue(&q, dep_tc(d));
+ if (ret == DEPT_BFS_CONTINUE_REV)
+ extend_queue_rev(&q, dep_fc(d));
+ }
+
+ while (!empty(&q))
+ dequeue(&q);
+}
+
+/*
+ * Main operations
+ * =====================================================================
+ * Add dependencies - Each new dependency is added into the graph and
+ * checked if it forms a circular dependency.
+ *
+ * Track waits - Waits are queued into the ring buffer for later use to
+ * generate appropriate dependencies with cross-event.
+ *
+ * Track event contexts(ecxt) - Event contexts are pushed into local
+ * stack for later use to generate appropriate dependencies with waits.
+ */
+
+static inline unsigned long cur_enirqf(void);
+static inline int cur_irq(void);
+static inline unsigned int cur_ctxt_id(void);
+
+static inline struct dept_stack *get_current_stack(void)
+{
+ struct dept_stack *s = dept_task()->stack;
+ return s ? get_stack(s) : NULL;
+}
+
+static inline void prepare_current_stack(void)
+{
+ struct dept_stack *s = dept_task()->stack;
+ /*
+ * The reference counter would be 1 when no one refers to the
+ * dept_stack.
+ */
+ const int expect_ref = 1;
+
+ /*
+ * The dept_stack is already ready.
+ */
+ if (s && !stack_refered(s, expect_ref)) {
+ s->nr = 0;
+ return;
+ }
+
+ if (s)
+ put_stack(s);
+
+ s = dept_task()->stack = new_stack();
+ if (!s)
+ return;
+
+ get_stack(s);
+ del_stack(s);
+}
+
+static void save_current_stack(int skip)
+{
+ struct dept_stack *s = dept_task()->stack;
+
+ if (!s)
+ return;
+ if (valid_stack(s))
+ return;
+
+ s->nr = stack_trace_save(s->raw, DEPT_MAX_STACK_ENTRY, skip);
+}
+
+static void finish_current_stack(void)
+{
+ struct dept_stack *s = dept_task()->stack;
+ /*
+ * The reference counter would be 1 when no one refers to the
+ * dept_stack.
+ */
+ const int expect_ref = 1;
+
+ if (s && stack_refered(s, expect_ref))
+ save_current_stack(2);
+}
+
+/*
+ * FIXME: For now, disable Lockdep while Dept is working.
+ *
+ * Both Lockdep and Dept report it on a deadlock detection using
+ * printk taking the risk of another deadlock that might be caused by
+ * locks of console or printk between inside and outside of them.
+ *
+ * For Dept, it's no problem since multiple reports are allowed. But it
+ * would be a bad idea for Lockdep since it will stop even on a singe
+ * report. So we need to prevent Lockdep from its reporting the risk
+ * Dept would take when reporting something.
+ */
+#include <linux/lockdep.h>
+
+void dept_off(void)
+{
+ dept_task()->recursive++;
+ lockdep_off();
+}
+
+void dept_on(void)
+{
+ dept_task()->recursive--;
+ lockdep_on();
+}
+
+static inline unsigned long dept_enter(void)
+{
+ unsigned long flags;
+ raw_local_irq_save(flags);
+ dept_off();
+ prepare_current_stack();
+ return flags;
+}
+
+static inline void dept_exit(unsigned long flags)
+{
+ finish_current_stack();
+ dept_on();
+ raw_local_irq_restore(flags);
+}
+
+/*
+ * NOTE: Must be called with dept_lock held.
+ */
+static struct dept_dep *__add_dep(struct dept_ecxt *e,
+ struct dept_wait *w)
+{
+ struct dept_dep *d;
+
+ if (!valid_class(e->class) || !valid_class(w->class))
+ return NULL;
+
+ if (dep(e->class, w->class))
+ return NULL;
+
+ d = new_dep();
+ if (unlikely(!d))
+ return NULL;
+
+ d->ecxt = get_ecxt(e);
+ d->wait = get_wait(w);
+
+ /*
+ * Add the dependency into hash and graph.
+ */
+ hash_add_dep(d);
+ list_add(&d->dep_node , &dep_fc(d)->dep_head );
+ list_add(&d->dep_rev_node, &dep_tc(d)->dep_rev_head);
+ return d;
+}
+
+static enum bfs_ret cb_check_dl(struct dept_dep *d,
+ void *in, void **out)
+{
+ struct dept_dep *new = (struct dept_dep *)in;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d) {
+ dep_tc(new)->bfs_parent = dep_fc(new);
+
+ if (dep_tc(new) != dep_fc(new))
+ return DEPT_BFS_CONTINUE;
+
+ /*
+ * AA circle does not make additional deadlock. We don't
+ * have to continue this BFS search.
+ */
+ print_circle(dep_tc(new));
+ return DEPT_BFS_DONE;
+ }
+
+ /*
+ * Allow multiple reports.
+ */
+ if (dep_tc(d) == dep_fc(new))
+ print_circle(dep_tc(new));
+
+ return DEPT_BFS_CONTINUE;
+}
+
+/*
+ * This function is actually in charge of reporting.
+ */
+static inline void check_dl(struct dept_dep *d)
+{
+ bfs(dep_tc(d), cb_check_dl, (void *)d, NULL);
+}
+
+/*
+ * Should keep the search until the end even if it encounters
+ * iwait_dist <= bfs_dist on the way because there might be
+ * DEPT_IWAIT_UNKNOWN or something here and there which should be filled
+ * with a valid one.
+ *
+ * XXX: The search by the branch can stop when it encounters
+ * iwait_dist <= bfs_dist, if it's proved that there are no need to go
+ * in the case.
+ */
+static enum bfs_ret cb_prop_iwait(struct dept_dep *d,
+ void *in, void **out)
+{
+ int irq = *(int *)in;
+ struct dept_class *fc;
+ struct dept_class *tc;
+ struct dept_dep *new;
+
+ if (DEPT_WARN_ON(!out))
+ return DEPT_BFS_DONE;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d)
+ return DEPT_BFS_CONTINUE;
+
+ fc = dep_fc(d);
+ tc = dep_tc(d);
+
+ if (tc->iwait_dist[irq] <= tc->bfs_dist)
+ return DEPT_BFS_CONTINUE;
+
+ tc->iwait_dist[irq] = tc->bfs_dist;
+ WRITE_ONCE(tc->iwait[irq], fc->iwait[irq]);
+
+ if (!tc->iecxt[irq])
+ return DEPT_BFS_CONTINUE;
+
+ /*
+ * Add irq dependency to form a complete circle.
+ */
+ new = __add_dep(tc->iecxt[irq], tc->iwait[irq]);
+ if (!new)
+ return DEPT_BFS_CONTINUE;
+
+ *out = new;
+ return DEPT_BFS_DONE;
+}
+
+/*
+ * Should keep the search until the end even if it encounters another
+ * iwait on the way because there might be the iwait here and there by
+ * any chance which should be cleaned as well.
+ *
+ * XXX: The search by the branch can stop when it encounters another
+ * iwait, if it's proved that there are no more the iwait in the case.
+ */
+static enum bfs_ret cb_clean_iwait(struct dept_dep *d, void *in, void **out)
+{
+ struct dept_class *c = (struct dept_class *)in;
+ struct dept_class *tc;
+ int i;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d)
+ return DEPT_BFS_CONTINUE;
+
+ tc = dep_tc(d);
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ if (!tc->iwait[i] || tc->iwait[i] != c->iwait[i])
+ continue;
+ WRITE_ONCE(tc->iwait[i], DEPT_IWAIT_UNKNOWN);
+ tc->iwait_dist[i] = INT_MAX;
+ }
+
+ return DEPT_BFS_CONTINUE;
+}
+
+/*
+ * Should perform cleaning iwait even if c is not a root of the iwait
+ * because c might be a bridge for the iwait to propagate.
+ */
+static void clean_iwait(struct dept_class *root, struct dept_class *c)
+{
+ bfs(root, cb_clean_iwait, (void *)c, NULL);
+}
+
+static LIST_HEAD(staleies);
+static LIST_HEAD(staleiws);
+
+static void stale_iecxt_iwait(struct dept_dep *d, int irq)
+{
+ struct dept_ecxt *ie = d->ecxt;
+ struct dept_wait *iw = d->wait;
+ struct dept_class *iwroot = iw->class;
+ struct dept_staleie *sie;
+ struct dept_staleiw *siw;
+ struct dept_class onetime_c;
+ int i;
+
+ /*
+ * iwroot should be root of the iwait.
+ */
+ DEPT_WARN_ON(iwroot->iwait_dist[irq] != 0);
+
+ /*
+ * ie and iw should have one common irq flag at least.
+ */
+ DEPT_WARN_ON(!(ie->enirqf & iw->irqf));
+
+ for (i = 0; i < DEPT_IRQS_NR; i++)
+ onetime_c.iwait[i] = (i == irq) ? iw : NULL;
+ clean_iwait(iwroot, &onetime_c);
+
+ sie = new_staleie();
+ if (likely(sie)) {
+ sie->ip = ie->enirq_ip[irq];
+ sie->irq = irq;
+ hash_add_staleie(sie);
+ list_add(&sie->all_node, &staleies);
+ }
+ ie->enirqf &= ~(1UL << irq);
+ put_ecxt(ie);
+ WRITE_ONCE(dep_fc(d)->iecxt[irq], NULL);
+
+ siw = new_staleiw();
+ if (likely(siw)) {
+ siw->ip = iw->irq_ip[irq];
+ siw->irq = irq;
+ hash_add_staleiw(siw);
+ list_add(&siw->all_node, &staleiws);
+ }
+ iw->irqf &= ~(1UL << irq);
+ put_wait(iw);
+ WRITE_ONCE(dep_tc(d)->iwait[irq], NULL);
+}
+
+/*
+ * Should start from a class that has 0 of iwait_dist.
+ */
+static void propagate_iwait(struct dept_class *c, int irq)
+{
+ struct dept_dep *new;
+ do {
+ new = NULL;
+ bfs(c, cb_prop_iwait, (void *)&irq, (void **)&new);
+
+ /*
+ * Deadlock detected. Let check_dl() report it.
+ */
+ if (new) {
+ check_dl(new);
+ stale_iecxt_iwait(new, irq);
+ }
+ } while (new);
+}
+
+static enum bfs_ret cb_find_iwait(struct dept_dep *d, void *in, void **out)
+{
+ int irq = *(int *)in;
+ struct dept_wait *w;
+
+ if (DEPT_WARN_ON(!out))
+ return DEPT_BFS_DONE;
+
+ /*
+ * initial condition for this BFS search
+ */
+ if (!d)
+ return DEPT_BFS_CONTINUE_REV;
+
+ w = dep_fc(d)->iwait[irq];
+ if (!w || w == DEPT_IWAIT_UNKNOWN)
+ return DEPT_BFS_CONTINUE_REV;
+
+ *out = w;
+ return DEPT_BFS_DONE;
+}
+
+static struct dept_wait *find_iwait(struct dept_class *c, int irq)
+{
+ struct dept_wait *w = NULL;
+ bfs(c, cb_find_iwait, (void *)&irq, (void **)&w);
+ return w;
+}
+
+static void add_iecxt(struct dept_ecxt *e, int irq, bool stack)
+{
+ /*
+ * This access is safe since we ensure e->class has set locally.
+ */
+ struct dept_class *c = e->class;
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(c->iecxt[irq]))
+ return;
+
+ if (unlikely(staleie(dt->enirq_ip[irq], irq)))
+ return;
+
+ if (unlikely(!dept_lock()))
+ return;
+
+ if (c->iecxt[irq])
+ goto unlock;
+
+ if (unlikely(staleie(dt->enirq_ip[irq], irq)))
+ goto unlock;
+
+ e->enirqf |= (1UL << irq);
+
+ WRITE_ONCE(c->iecxt[irq], get_ecxt(e));
+
+ /*
+ * Should be NULL since it's the first time that these
+ * enirq_{ip,stack}[irq] have ever set.
+ */
+ DEPT_WARN_ON(e->enirq_ip[irq]);
+ DEPT_WARN_ON(e->enirq_stack[irq]);
+
+ e->enirq_ip[irq] = dt->enirq_ip[irq];
+ if (stack)
+ e->enirq_stack[irq] = get_current_stack();
+
+ if (c->iwait[irq] == DEPT_IWAIT_UNKNOWN) {
+ struct dept_wait *w = find_iwait(c, irq);
+
+ /*
+ * Deadlock detected. Let propagate_iwait() report it.
+ */
+ if (w)
+ propagate_iwait(w->class, irq);
+ } else if (c->iwait[irq]) {
+ /*
+ * Add irq dependency to form a complete circle.
+ */
+ struct dept_dep *d = __add_dep(e, c->iwait[irq]);
+
+ /*
+ * Deadlock detected. Let check_dl() report it.
+ */
+ if (d) {
+ check_dl(d);
+ stale_iecxt_iwait(d, irq);
+ }
+ }
+unlock:
+ dept_unlock();
+}
+
+static void add_iwait(struct dept_wait *w, int irq)
+{
+ struct dept_class *c = w->class;
+ struct dept_wait *iw;
+
+ iw = READ_ONCE(c->iwait[irq]);
+ if (iw && READ_ONCE(iw->class) == c)
+ return;
+
+ if (unlikely(staleiw(w->irq_ip[irq], irq)))
+ return;
+
+ if (unlikely(!dept_lock()))
+ return;
+
+ iw = c->iwait[irq];
+ if (iw && iw->class == c)
+ goto unlock;
+
+ if (unlikely(staleiw(w->irq_ip[irq], irq)))
+ goto unlock;
+
+ w->irqf |= (1UL << irq);
+
+ c->iwait_dist[irq] = 0;
+ WRITE_ONCE(c->iwait[irq], get_wait(w));
+
+ /*
+ * Should be NULL since it's the first time that these
+ * irq_{ip,stack}[irq] have ever set.
+ */
+ DEPT_WARN_ON(w->irq_ip[irq]);
+ DEPT_WARN_ON(w->irq_stack[irq]);
+
+ w->irq_ip[irq] = w->wait_ip;
+ w->irq_stack[irq] = get_current_stack();
+
+ if (c->iecxt[irq]) {
+ /*
+ * Add irq dependency to form a complete circle.
+ */
+ struct dept_dep *d = __add_dep(c->iecxt[irq], w);
+
+ /*
+ * Deadlock detected. Let check_dl() report it.
+ */
+ if (d) {
+ check_dl(d);
+ stale_iecxt_iwait(d, irq);
+ }
+ }
+ propagate_iwait(c, irq);
+unlock:
+ dept_unlock();
+}
+
+static inline struct dept_wait_hist *hist(int pos)
+{
+ struct dept_task *dt = dept_task();
+ return dt->wait_hist + (pos % DEPT_MAX_WAIT_HIST);
+}
+
+static inline int hist_pos_next(void)
+{
+ struct dept_task *dt = dept_task();
+ return dt->wait_hist_pos % DEPT_MAX_WAIT_HIST;
+}
+
+static inline void hist_advance(void)
+{
+ struct dept_task *dt = dept_task();
+ dt->wait_hist_pos++;
+ dt->wait_hist_pos %= DEPT_MAX_WAIT_HIST;
+}
+
+static inline struct dept_wait_hist *new_hist(void)
+{
+ struct dept_wait_hist *wh = hist(hist_pos_next());
+ hist_advance();
+ return wh;
+}
+
+static void add_hist(struct dept_wait *w, unsigned int wg, unsigned int ctxt_id)
+{
+ struct dept_wait_hist *wh = new_hist();
+
+ if (likely(wh->wait))
+ put_wait(wh->wait);
+
+ wh->wait = get_wait(w);
+ wh->wgen = wg;
+ wh->ctxt_id = ctxt_id;
+}
+
+static void add_dep(struct dept_ecxt *e, struct dept_wait *w)
+{
+ struct dept_dep *d;
+ int i;
+
+ if (dep(e->class, w->class))
+ return;
+
+ if (unlikely(!dept_lock()))
+ return;
+
+ d = __add_dep(e, w);
+ if (d) {
+ check_dl(d);
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ /*
+ * There's no iwait to propagate in the 'from'.
+ */
+ if (!e->class->iwait[i])
+ continue;
+ /*
+ * This case has been already handled by add_iwait().
+ */
+ if (w->class->iwait[i])
+ continue;
+
+ propagate_iwait(e->class->iwait[i]->class, i);
+ }
+ }
+ dept_unlock();
+}
+
+static atomic_t wgen = ATOMIC_INIT(1);
+
+static void add_wait(struct dept_class *c, unsigned long ip,
+ const char *w_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_wait *w;
+ unsigned int wg = 0U;
+ int irq;
+ int i;
+
+ w = new_wait();
+ if (unlikely(!w))
+ return;
+
+ WRITE_ONCE(w->class, get_class(c));
+ w->wait_ip = ip;
+ w->wait_fn = w_fn;
+ w->wait_stack = get_current_stack();
+
+ irq = cur_irq();
+ if (irq < DEPT_IRQS_NR)
+ add_iwait(w, irq);
+
+ /*
+ * Avoid adding dependency between user aware nested ecxt and
+ * wait.
+ */
+ for (i = dt->ecxt_held_pos - 1; i >= 0; i--) {
+ struct dept_ecxt_held *eh;
+ eh = dt->ecxt_held + i;
+ if (eh->ecxt->class != c || eh->nest == ne)
+ break;
+ }
+
+ for (; i >= 0; i--) {
+ struct dept_ecxt_held *eh;
+ eh = dt->ecxt_held + i;
+ add_dep(eh->ecxt, w);
+ }
+
+ if (!wait_refered(w, 1) && !rich_stack) {
+ if (w->wait_stack)
+ put_stack(w->wait_stack);
+ w->wait_stack = NULL;
+ }
+
+ /*
+ * Avoid zero wgen.
+ */
+ wg = atomic_inc_return(&wgen) ?: atomic_inc_return(&wgen);
+ add_hist(w, wg, cur_ctxt_id());
+
+ del_wait(w);
+}
+
+static void add_ecxt(void *obj, struct dept_class *c, unsigned long ip,
+ const char *c_fn, const char *e_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_ecxt_held *eh;
+ struct dept_ecxt *e;
+ unsigned long irqf;
+ int irq;
+
+ if (DEPT_WARN_ON(dt->ecxt_held_pos == DEPT_MAX_ECXT_HELD))
+ return;
+
+ e = new_ecxt();
+ if (unlikely(!e))
+ return;
+
+ e->class = get_class(c);
+ e->ecxt_ip = ip;
+ e->ecxt_stack = ip && rich_stack ? get_current_stack() : NULL;
+ e->event_fn = e_fn;
+ e->ecxt_fn = c_fn;
+
+ eh = dt->ecxt_held + (dt->ecxt_held_pos++);
+ eh->ecxt = get_ecxt(e);
+ eh->key = (unsigned long)obj;
+ eh->wgen = atomic_read(&wgen);
+ eh->nest = ne;
+
+ irqf = cur_enirqf();
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR)
+ add_iecxt(e, irq, false);
+
+ del_ecxt(e);
+}
+
+static int find_ecxt_pos(unsigned long key, bool newfirst)
+{
+ struct dept_task *dt = dept_task();
+ int i;
+
+ if (newfirst) {
+ for (i = dt->ecxt_held_pos - 1; i >= 0; i--)
+ if (dt->ecxt_held[i].key == key)
+ return i;
+ } else {
+ for (i = 0; i < dt->ecxt_held_pos; i++)
+ if (dt->ecxt_held[i].key == key)
+ return i;
+ }
+ return -1;
+}
+
+static void pop_ecxt(void *obj)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long key = (unsigned long)obj;
+ int pos;
+ int i;
+
+ /*
+ * TODO: WARN on pos == -1.
+ */
+ pos = find_ecxt_pos(key, true);
+ if (pos == -1)
+ return;
+
+ put_ecxt(dt->ecxt_held[pos].ecxt);
+ dt->ecxt_held_pos--;
+
+ for (i = pos; i < dt->ecxt_held_pos; i++)
+ dt->ecxt_held[i] = dt->ecxt_held[i + 1];
+}
+
+static inline bool good_hist(struct dept_wait_hist *wh, unsigned int wg)
+{
+ return wh->wait != NULL && before(wg, wh->wgen);
+}
+
+/*
+ * Binary-search the ring buffer for the earliest valid wait.
+ */
+static int find_hist_pos(unsigned int wg)
+{
+ int oldest;
+ int l;
+ int r;
+ int pos;
+
+ oldest = hist_pos_next();
+ if (unlikely(good_hist(hist(oldest), wg))) {
+ DEPT_WARN_ONCE("Need to expand the ring buffer.\n");
+ return oldest;
+ }
+
+ l = oldest + 1;
+ r = oldest + DEPT_MAX_WAIT_HIST - 1;
+ for (pos = (l + r) / 2; l <= r; pos = (l + r) / 2) {
+ struct dept_wait_hist *p = hist(pos - 1);
+ struct dept_wait_hist *wh = hist(pos);
+
+ if (!good_hist(p, wg) && good_hist(wh, wg))
+ return pos % DEPT_MAX_WAIT_HIST;
+ if (good_hist(wh, wg))
+ r = pos - 1;
+ else
+ l = pos + 1;
+ }
+ return -1;
+}
+
+static void do_event(void *obj, struct dept_class *c, unsigned int wg,
+ unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_wait_hist *wh;
+ struct dept_ecxt_held *eh;
+ unsigned long key = (unsigned long)obj;
+ unsigned int ctxt_id;
+ int end;
+ int pos;
+ int i;
+
+ /*
+ * The event was triggered before wait.
+ */
+ if (!wg)
+ return;
+
+ pos = find_ecxt_pos(key, false);
+ if (pos == -1)
+ return;
+
+ eh = dt->ecxt_held + pos;
+ eh->ecxt->event_ip = ip;
+ eh->ecxt->event_stack = get_current_stack();
+
+ /*
+ * The ecxt already has done what it needs.
+ */
+ if (!before(wg, eh->wgen))
+ return;
+
+ pos = find_hist_pos(wg);
+ if (pos == -1)
+ return;
+
+ ctxt_id = cur_ctxt_id();
+ end = hist_pos_next();
+ end = end > pos ? end : end + DEPT_MAX_WAIT_HIST;
+ for (wh = hist(pos); pos < end; wh = hist(++pos)) {
+ if (wh->ctxt_id == ctxt_id)
+ add_dep(eh->ecxt, wh->wait);
+ if (!before(wh->wgen, eh->wgen))
+ break;
+ }
+
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ if (before(dt->wgen_enirq[i], wg))
+ continue;
+ add_iecxt(eh->ecxt, i, false);
+ }
+}
+
+static void del_dep_rcu(struct rcu_head *rh)
+{
+ struct dept_dep *d = container_of(rh, struct dept_dep, rh);
+
+ preempt_disable();
+ del_dep(d);
+ preempt_enable();
+}
+
+/*
+ * NOTE: Must be called with dept_lock held.
+ */
+static void disconnect_class(struct dept_class *c)
+{
+ struct dept_dep *d, *n;
+ int i;
+
+ clean_iwait(c, c);
+ for (i = 0; i < DEPT_IRQS_NR; i++) {
+ struct dept_ecxt *e = c->iecxt[i];
+ struct dept_wait *w = c->iwait[i];
+ if (e)
+ put_ecxt(e);
+ if (w && w->class == c)
+ put_wait(w);
+ WRITE_ONCE(c->iecxt[i], NULL);
+ WRITE_ONCE(c->iwait[i], NULL);
+ }
+
+ list_for_each_entry_safe(d, n, &c->dep_head, dep_node) {
+ list_del_rcu(&d->dep_node);
+ list_del_rcu(&d->dep_rev_node);
+ hash_del_dep(d);
+ call_rcu(&d->rh, del_dep_rcu);
+ }
+
+ list_for_each_entry_safe(d, n, &c->dep_rev_head, dep_rev_node) {
+ list_del_rcu(&d->dep_node);
+ list_del_rcu(&d->dep_rev_node);
+ hash_del_dep(d);
+ call_rcu(&d->rh, del_dep_rcu);
+ }
+}
+
+/*
+ * IRQ context control
+ * =====================================================================
+ * Whether a wait is in {hard,soft}-IRQ context or whether
+ * {hard,soft}-IRQ has been enabled on the way to an event is very
+ * important to check dependency. All those things should be tracked.
+ */
+
+static inline unsigned long cur_enirqf(void)
+{
+ struct dept_task *dt = dept_task();
+ int he = dt->hardirqs_enabled;
+ int se = dt->softirqs_enabled;
+ if (he)
+ return DEPT_HIRQF | (se ? DEPT_SIRQF : 0UL);
+ return 0UL;
+}
+
+static inline int cur_irq(void)
+{
+ if (lockdep_softirq_context(current))
+ return DEPT_SIRQ;
+ if (lockdep_hardirq_context())
+ return DEPT_HIRQ;
+ return DEPT_IRQS_NR;
+}
+
+static inline unsigned int cur_ctxt_id(void)
+{
+ struct dept_task *dt = dept_task();
+ int irq = cur_irq();
+
+ /*
+ * Normal process context
+ */
+ if (irq == DEPT_IRQS_NR)
+ return 0U;
+
+ return dt->irq_id[irq] | (1UL << irq);
+}
+
+static void enirq_transition(int irq)
+{
+ struct dept_task *dt = dept_task();
+ int i;
+
+ /*
+ * READ wgen >= wgen of an event with IRQ enabled has been
+ * observed on the way to the event means, the IRQ can cut in
+ * within the ecxt. Used for cross-event detection.
+ *
+ * wait context event context(ecxt)
+ * ------------ -------------------
+ * wait event
+ * WRITE wgen
+ * observe IRQ enabled
+ * READ wgen
+ * keep the wgen locally
+ *
+ * on the event
+ * check the local wgen
+ */
+ dt->wgen_enirq[irq] = atomic_read(&wgen);
+
+ for (i = dt->ecxt_held_pos - 1; i >= 0; i--) {
+ struct dept_ecxt_held *eh;
+ eh = dt->ecxt_held + i;
+ add_iecxt(eh->ecxt, irq, true);
+ }
+}
+
+static void enirq_update(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long irqf;
+ unsigned long prev;
+ int irq;
+
+ prev = dt->eff_enirqf;
+ irqf = cur_enirqf();
+ dt->eff_enirqf = irqf;
+
+ /*
+ * Do enirq_transition() only on an OFF -> ON transition.
+ */
+ for_each_set_bit(irq, &irqf, DEPT_IRQS_NR) {
+ unsigned long flags;
+
+ if (prev & (1UL << irq))
+ continue;
+
+ flags = dept_enter();
+ dt->enirq_ip[irq] = ip;
+ enirq_transition(irq);
+ dept_exit(flags);
+ }
+}
+
+/*
+ * Ensure it has been called on OFF -> ON transition.
+ */
+void dept_enable_softirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(early_boot_irqs_disabled))
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->softirqs_enabled = true;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it has been called on OFF -> ON transition.
+ */
+void dept_enable_hardirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(early_boot_irqs_disabled))
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->hardirqs_enabled = true;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it has been called on ON -> OFF transition.
+ */
+void dept_disable_softirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->softirqs_enabled = false;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it has been called on ON -> OFF transition.
+ */
+void dept_disable_hardirq(unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (DEPT_WARN_ON(!irqs_disabled()))
+ return;
+
+ dt->hardirqs_enabled = false;
+ enirq_update(ip);
+}
+
+/*
+ * Ensure it's the outmost softirq context.
+ */
+void dept_softirq_enter(void)
+{
+ struct dept_task *dt = dept_task();
+ dt->irq_id[DEPT_SIRQ] += (1UL << DEPT_IRQS_NR);
+}
+
+/*
+ * Ensure it's the outmost hardirq context.
+ */
+void dept_hardirq_enter(void)
+{
+ struct dept_task *dt = dept_task();
+ dt->irq_id[DEPT_HIRQ] += (1UL << DEPT_IRQS_NR);
+}
+
+/*
+ * Dept API
+ * =====================================================================
+ * Main Dept APIs.
+ */
+
+static inline void clean_classes_cache(struct dept_key *k)
+{
+ int i;
+
+ for (i = 0; i < DEPT_MAX_SUBCLASSES_CACHE; i++)
+ k->classes[i] = NULL;
+}
+
+void dept_map_init(struct dept_map *m, struct dept_key *k, int sub,
+ const char *n)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ if(DEPT_WARN_ON(sub < 0 || sub >= DEPT_MAX_SUBCLASSES_USR)) {
+ m->nocheck = true;
+ goto exit;
+ }
+
+ if (m->keys != k)
+ m->keys = k;
+ clean_classes_cache(&m->keys_local);
+
+ m->sub_usr = sub;
+ m->name = n;
+ m->wgen = 0U;
+ m->nocheck = false;
+exit:
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_map_init);
+
+void dept_map_reinit(struct dept_map *m)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ clean_classes_cache(&m->keys_local);
+ m->wgen = 0U;
+ m->nocheck = false;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_map_reinit);
+
+void dept_map_nocheck(struct dept_map *m)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ m->nocheck = true;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_map_nocheck);
+
+static LIST_HEAD(classes);
+
+static inline bool within(const void *addr, void *start, unsigned long size)
+{
+ return addr >= start && addr < start + size;
+}
+
+static void free_staleie_staleiw_range(void *start, unsigned int sz)
+{
+ struct dept_staleie *sie, *sien;
+ struct dept_staleiw *siw, *siwn;
+
+ list_for_each_entry_safe(sie, sien, &staleies, all_node)
+ if (within((void *)sie->ip, start, sz))
+ hash_del_staleie(sie);
+
+ list_for_each_entry_safe(siw, siwn, &staleiws, all_node)
+ if (within((void *)siw->ip, start, sz))
+ hash_del_staleiw(siw);
+}
+
+void dept_free_range(void *start, unsigned int sz)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_class *c, *n;
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * dept_free_range() should not fail.
+ *
+ * FIXME: Should be fixed if dept_free_range() causes deadlock
+ * with dept_lock().
+ */
+ while (unlikely(!dept_lock()));
+
+ /*
+ * Free all staleies and staleiws within the range.
+ */
+ free_staleie_staleiw_range(start, sz);
+
+ list_for_each_entry_safe(c, n, &classes, all_node) {
+ if (!within((void *)c->key, start, sz) &&
+ !within(c->name, start, sz))
+ continue;
+
+ hash_del_class(c);
+ disconnect_class(c);
+ list_del(&c->all_node);
+ invalidate_class(c);
+
+ /*
+ * Actual deletion will happen on the rcu callback
+ * that has been added in disconnect_class().
+ */
+ del_class(c);
+ }
+ dept_unlock();
+ dept_exit(flags);
+
+ /*
+ * Wait until even lockless hash_lookup_class() for the class
+ * returns NULL.
+ */
+ might_sleep();
+ synchronize_rcu();
+}
+
+static inline int map_sub(struct dept_map *m, int e)
+{
+ return m->sub_usr + e * DEPT_MAX_SUBCLASSES_USR;
+}
+
+static struct dept_class *check_new_class(struct dept_key *local,
+ struct dept_key *k, int sub,
+ const char *n)
+{
+ struct dept_class *c = NULL;
+
+ if (DEPT_WARN_ON(sub >= DEPT_MAX_SUBCLASSES))
+ return NULL;
+
+ if (DEPT_WARN_ON(!k))
+ return NULL;
+
+ if (sub < DEPT_MAX_SUBCLASSES_CACHE)
+ c = READ_ONCE(local->classes[sub]);
+
+ if (c)
+ return c;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (c)
+ goto caching;
+
+ if (unlikely(!dept_lock()))
+ return NULL;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (unlikely(c))
+ goto unlock;
+
+ c = new_class();
+ if (unlikely(!c))
+ goto unlock;
+
+ c->name = n;
+ c->sub = sub;
+ c->key = (unsigned long)(k->subkeys + sub);
+ hash_add_class(c);
+ list_add(&c->all_node, &classes);
+unlock:
+ dept_unlock();
+caching:
+ if (sub < DEPT_MAX_SUBCLASSES_CACHE && c)
+ WRITE_ONCE(local->classes[sub], c);
+
+ return c;
+}
+
+void dept_wait(struct dept_map *m, unsigned long w_f, unsigned long ip,
+ const char *w_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int e;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * Be as conservative as possible. In case of mulitple waits for
+ * a single dept_map, we are going to keep only the last wait's
+ * wgen for simplicity - keeping all wgens seems overengineering.
+ *
+ * Of course, it might cause missing some dependencies that
+ * would rarely, probabily never, happen but it helps avoid
+ * false positive report.
+ */
+ for_each_set_bit(e, &w_f, DEPT_MAX_SUBCLASSES_EVT) {
+ struct dept_class *c;
+ struct dept_key *k;
+
+ k = m->keys ?: &m->keys_local;
+ c = check_new_class(&m->keys_local, k,
+ map_sub(m, e), m->name);
+ if (!c)
+ continue;
+
+ add_wait(c, ip, w_fn, ne);
+ }
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_wait);
+
+void dept_ecxt_enter(struct dept_map *m, unsigned long e_f, unsigned long ip,
+ const char *c_fn, const char *e_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int e;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ for_each_set_bit(e, &e_f, DEPT_MAX_SUBCLASSES_EVT) {
+ struct dept_class *c;
+ struct dept_key *k;
+
+ k = m->keys ?: &m->keys_local;
+ c = check_new_class(&m->keys_local, k,
+ map_sub(m, e), m->name);
+ if (!c)
+ continue;
+
+ add_ecxt((void *)m, c, ip, c_fn, e_fn, ne);
+ }
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_ecxt_enter);
+
+void dept_asked_event(struct dept_map *m)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ unsigned int wg;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * Avoid zero wgen.
+ */
+ wg = atomic_inc_return(&wgen) ?: atomic_inc_return(&wgen);
+ WRITE_ONCE(m->wgen, wg);
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_asked_event);
+
+void dept_event(struct dept_map *m, unsigned long e_f, unsigned long ip,
+ const char *e_fn)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int e;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ for_each_set_bit(e, &e_f, DEPT_MAX_SUBCLASSES_EVT) {
+ struct dept_class *c;
+ struct dept_key *k;
+
+ k = m->keys ?: &m->keys_local;
+ c = check_new_class(&m->keys_local, k,
+ map_sub(m, e), m->name);
+ if (!c)
+ continue;
+
+ add_ecxt((void *)m, c, 0UL, NULL, e_fn, 0);
+ do_event((void *)m, c, READ_ONCE(m->wgen), ip);
+ pop_ecxt((void *)m);
+ }
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_event);
+
+void dept_ecxt_exit(struct dept_map *m, unsigned long ip)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (m->nocheck)
+ return;
+
+ flags = dept_enter();
+ pop_ecxt((void *)m);
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_ecxt_exit);
+
+struct dept_map *dept_top_map(void)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_map *m;
+ unsigned long flags;
+ int pos;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return NULL;
+
+ flags = dept_enter();
+ pos = dt->ecxt_held_pos;
+ m = pos ? (struct dept_map *)dt->ecxt_held[pos - 1].key : NULL;
+ dept_exit(flags);
+
+ return m;
+}
+EXPORT_SYMBOL_GPL(dept_top_map);
+
+void dept_warn_on(bool cond)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+ DEPT_WARN_ON(cond);
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_warn_on);
+
+void dept_task_exit(struct task_struct *t)
+{
+ struct dept_task *dt = &t->dept_task;
+ int i;
+
+ raw_local_irq_disable();
+
+ if (dt->stack)
+ put_stack(dt->stack);
+
+ for (i = 0; i < dt->ecxt_held_pos; i++)
+ put_ecxt(dt->ecxt_held[i].ecxt);
+
+ for (i = 0; i < DEPT_MAX_WAIT_HIST; i++)
+ if (dt->wait_hist[i].wait)
+ put_wait(dt->wait_hist[i].wait);
+
+ dept_off();
+
+ raw_local_irq_enable();
+}
+
+void dept_task_init(struct task_struct *t)
+{
+ memset(&t->dept_task, 0x0, sizeof(struct dept_task));
+}
+
+void dept_key_init(struct dept_key *k)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int sub;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * dept_key_init() should not fail.
+ *
+ * FIXME: Should be fixed if dept_key_init() causes deadlock
+ * with dept_lock().
+ */
+ while (unlikely(!dept_lock()));
+
+ for (sub = 0; sub < DEPT_MAX_SUBCLASSES; sub++) {
+ struct dept_class *c;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (!c)
+ continue;
+
+ DEPT_STOP("The class(%s/%d) has not been removed.\n",
+ c->name, sub);
+ break;
+ }
+
+ dept_unlock();
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_key_init);
+
+void dept_key_destroy(struct dept_key *k)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ int sub;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * dept_key_destroy() should not fail.
+ *
+ * FIXME: Should be fixed if dept_key_destroy() causes deadlock
+ * with dept_lock().
+ */
+ while (unlikely(!dept_lock()));
+
+ for (sub = 0; sub < DEPT_MAX_SUBCLASSES; sub++) {
+ struct dept_class *c;
+
+ c = class((unsigned long)k->subkeys + sub);
+ if (!c)
+ continue;
+
+ hash_del_class(c);
+ disconnect_class(c);
+ list_del(&c->all_node);
+ invalidate_class(c);
+
+ /*
+ * Actual deletion will happen on the rcu callback
+ * that has been added in disconnect_class().
+ */
+ del_class(c);
+ }
+
+ dept_unlock();
+ dept_exit(flags);
+
+ /*
+ * Wait until even lockless hash_lookup_class() for the class
+ * returns NULL.
+ */
+ might_sleep();
+ synchronize_rcu();
+}
+EXPORT_SYMBOL_GPL(dept_key_destroy);
+
+static void move_llist(struct llist_head *to, struct llist_head *from)
+{
+ struct llist_node *first = llist_del_all(from);
+ struct llist_node *last;
+
+ if (!first)
+ return;
+
+ for (last = first; last->next; last = last->next);
+ llist_add_batch(first, last, to);
+}
+
+static void migrate_per_cpu_pool(void)
+{
+ const int boot_cpu = 0;
+ int i;
+
+ /*
+ * The boot CPU has been using the temperal local pool so far.
+ * From now on that per_cpu areas have been ready, use the
+ * per_cpu local pool instead.
+ */
+ DEPT_WARN_ON(smp_processor_id() != boot_cpu);
+ for (i = 0; i < OBJECT_NR; i++) {
+ struct llist_head *from;
+ struct llist_head *to;
+
+ from = &pool[i].boot_pool;
+ to = per_cpu_ptr(pool[i].lpool, boot_cpu);
+ move_llist(to, from);
+ }
+}
+
+#define B2KB(B) ((B) / 1024)
+
+/*
+ * Should be called after setup_per_cpu_areas() and before no non-boot
+ * CPUs have been on.
+ */
+void __init dept_init(void)
+{
+ size_t mem_total = 0;
+
+ local_irq_disable();
+ dept_per_cpu_ready = 1;
+ migrate_per_cpu_pool();
+ local_irq_enable();
+
+#define OBJECT(id, nr) mem_total += sizeof(struct dept_##id) * nr;
+ #include "dept_object.h"
+#undef OBJECT
+#define HASH(id, bits) mem_total += sizeof(struct hlist_head) * (1UL << bits);
+ #include "dept_hash.h"
+#undef HASH
+
+ printk("DEPendency Tracker: Copyright (c) 2020 LG Electronics, Inc., Byungchul Park\n");
+ printk("... DEPT_MAX_STACK_ENTRY: %d\n", DEPT_MAX_STACK_ENTRY);
+ printk("... DEPT_MAX_WAIT_HIST : %d\n", DEPT_MAX_WAIT_HIST );
+ printk("... DEPT_MAX_ECXT_HELD : %d\n", DEPT_MAX_ECXT_HELD );
+ printk("... DEPT_MAX_SUBCLASSES : %d\n", DEPT_MAX_SUBCLASSES );
+#define OBJECT(id, nr) \
+ printk("... memory used by %s: %zu KB\n", \
+ #id, B2KB(sizeof(struct dept_##id) * nr));
+ #include "dept_object.h"
+#undef OBJECT
+#define HASH(id, bits) \
+ printk("... hash list head used by %s: %zu KB\n", \
+ #id, B2KB(sizeof(struct hlist_head) * (1UL << bits)));
+ #include "dept_hash.h"
+#undef HASH
+ printk("... total memory used by objects and hashs: %zu KB\n", B2KB(mem_total));
+ printk("... per task memory footprint: %zu bytes\n", sizeof(struct dept_task));
+}
diff --git a/kernel/dependency/dept_hash.h b/kernel/dependency/dept_hash.h
new file mode 100644
index 0000000..4762977
--- /dev/null
+++ b/kernel/dependency/dept_hash.h
@@ -0,0 +1,11 @@
+/*
+ * HASH(id, bits)
+ *
+ * id : Id for the object of struct dept_##id.
+ * bits: 1UL << bits is the hash table size.
+ */
+
+HASH(dep , 12)
+HASH(class , 12)
+HASH(staleiw , 4)
+HASH(staleie , 4)
diff --git a/kernel/dependency/dept_object.h b/kernel/dependency/dept_object.h
new file mode 100644
index 0000000..3e8ba81
--- /dev/null
+++ b/kernel/dependency/dept_object.h
@@ -0,0 +1,14 @@
+/*
+ * OBJECT(id, nr)
+ *
+ * id: Id for the object of struct dept_##id.
+ * nr: # of the object that should be kept in the pool.
+ */
+
+OBJECT(dep ,1024 * 8)
+OBJECT(class ,1024 * 4)
+OBJECT(stack ,1024 * 32)
+OBJECT(ecxt ,1024 * 4)
+OBJECT(wait ,1024 * 32)
+OBJECT(staleiw ,16)
+OBJECT(staleie ,16)
diff --git a/kernel/exit.c b/kernel/exit.c
index b00a25b..187ee24 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -854,6 +854,7 @@ void __noreturn do_exit(long code)
exit_tasks_rcu_finish();
lockdep_free_task(tsk);
+ dept_task_exit(tsk);
do_task_dead();
}
diff --git a/kernel/fork.c b/kernel/fork.c
index d75a528f..8de918b 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -97,6 +97,7 @@
#include <linux/scs.h>
#include <linux/io_uring.h>
#include <linux/bpf.h>
+#include <linux/dept.h>
#include <asm/pgalloc.h>
#include <linux/uaccess.h>
@@ -2117,6 +2118,7 @@ static __latent_entropy struct task_struct *copy_process(
#ifdef CONFIG_LOCKDEP
lockdep_init_task(p);
#endif
+ dept_task_init(p);
#ifdef CONFIG_DEBUG_MUTEXES
p->blocked_on = NULL; /* not blocked yet */
diff --git a/kernel/module.c b/kernel/module.c
index 24dab04..bd9376d 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -2205,6 +2205,7 @@ static void free_module(struct module *mod)
/* Free lock-classes; relies on the preceding sync_rcu(). */
lockdep_free_key_range(mod->core_layout.base, mod->core_layout.size);
+ dept_free_range(mod->core_layout.base, mod->core_layout.size);
/* Finally, free the core (containing the module structure) */
module_memfree(mod->core_layout.base);
@@ -4174,6 +4175,7 @@ static int load_module(struct load_info *info, const char __user *uargs,
free_module:
/* Free lock-classes; relies on the preceding sync_rcu() */
lockdep_free_key_range(mod->core_layout.base, mod->core_layout.size);
+ dept_free_range(mod->core_layout.base, mod->core_layout.size);
module_deallocate(mod, info);
free_copy:
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 41f4709..a28c950 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -320,7 +320,7 @@ void __local_bh_disable_ip(unsigned long ip, unsigned int cnt)
* Were softirqs turned off above:
*/
if (softirq_count() == (cnt & SOFTIRQ_MASK))
- lockdep_softirqs_off(ip);
+ trace_softirqs_off_caller(ip);
raw_local_irq_restore(flags);
if (preempt_count() == cnt) {
@@ -341,7 +341,7 @@ static void __local_bh_enable(unsigned int cnt)
trace_preempt_on(CALLER_ADDR0, get_lock_parent_ip());
if (softirq_count() == (cnt & SOFTIRQ_MASK))
- lockdep_softirqs_on(_RET_IP_);
+ trace_softirqs_on_caller(_RET_IP_);
__preempt_count_sub(cnt);
}
@@ -368,7 +368,7 @@ void __local_bh_enable_ip(unsigned long ip, unsigned int cnt)
* Are softirqs going to be turned on now:
*/
if (softirq_count() == SOFTIRQ_DISABLE_OFFSET)
- lockdep_softirqs_on(ip);
+ trace_softirqs_on_caller(ip);
/*
* Keep preemption disabled until we are done with
* softirq processing:
diff --git a/kernel/trace/trace_preemptirq.c b/kernel/trace/trace_preemptirq.c
index f493804..19cafdfb 100644
--- a/kernel/trace/trace_preemptirq.c
+++ b/kernel/trace/trace_preemptirq.c
@@ -19,6 +19,18 @@
/* Per-cpu variable to prevent redundant calls when IRQs already off */
static DEFINE_PER_CPU(int, tracing_irq_cpu);
+void trace_softirqs_on_caller(unsigned long ip)
+{
+ lockdep_softirqs_on(ip);
+ dept_enable_softirq(ip);
+}
+
+void trace_softirqs_off_caller(unsigned long ip)
+{
+ lockdep_softirqs_off(ip);
+ dept_disable_softirq(ip);
+}
+
/*
* Like trace_hardirqs_on() but without the lockdep invocation. This is
* used in the low level entry code where the ordering vs. RCU is important
@@ -33,6 +45,7 @@ void trace_hardirqs_on_prepare(void)
tracer_hardirqs_on(CALLER_ADDR0, CALLER_ADDR1);
this_cpu_write(tracing_irq_cpu, 0);
}
+ dept_enable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_on_prepare);
NOKPROBE_SYMBOL(trace_hardirqs_on_prepare);
@@ -45,6 +58,7 @@ void trace_hardirqs_on(void)
tracer_hardirqs_on(CALLER_ADDR0, CALLER_ADDR1);
this_cpu_write(tracing_irq_cpu, 0);
}
+ dept_enable_hardirq(CALLER_ADDR0);
lockdep_hardirqs_on_prepare(CALLER_ADDR0);
lockdep_hardirqs_on(CALLER_ADDR0);
@@ -66,7 +80,7 @@ void trace_hardirqs_off_finish(void)
if (!in_nmi())
trace_irq_disable(CALLER_ADDR0, CALLER_ADDR1);
}
-
+ dept_disable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_off_finish);
NOKPROBE_SYMBOL(trace_hardirqs_off_finish);
@@ -81,6 +95,7 @@ void trace_hardirqs_off(void)
if (!in_nmi())
trace_irq_disable_rcuidle(CALLER_ADDR0, CALLER_ADDR1);
}
+ dept_disable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_off);
NOKPROBE_SYMBOL(trace_hardirqs_off);
@@ -93,6 +108,7 @@ __visible void trace_hardirqs_on_caller(unsigned long caller_addr)
tracer_hardirqs_on(CALLER_ADDR0, caller_addr);
this_cpu_write(tracing_irq_cpu, 0);
}
+ dept_enable_hardirq(CALLER_ADDR0);
lockdep_hardirqs_on_prepare(CALLER_ADDR0);
lockdep_hardirqs_on(CALLER_ADDR0);
@@ -110,6 +126,7 @@ __visible void trace_hardirqs_off_caller(unsigned long caller_addr)
if (!in_nmi())
trace_irq_disable_rcuidle(CALLER_ADDR0, caller_addr);
}
+ dept_disable_hardirq(CALLER_ADDR0);
}
EXPORT_SYMBOL(trace_hardirqs_off_caller);
NOKPROBE_SYMBOL(trace_hardirqs_off_caller);
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 14b89aa..309b275 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1233,6 +1233,26 @@ config DEBUG_PREEMPT
menu "Lock Debugging (spinlocks, mutexes, etc...)"
+config DEPT
+ bool "Dependency tracking"
+ depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT
+ select DEBUG_SPINLOCK
+ select DEBUG_MUTEXES
+ select DEBUG_RT_MUTEXES if RT_MUTEXES
+ select DEBUG_RWSEMS
+ select DEBUG_WW_MUTEX_SLOWPATH
+ select DEBUG_LOCK_ALLOC
+ select TRACE_IRQFLAGS
+ select STACKTRACE
+ select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86
+ select KALLSYMS
+ select KALLSYMS_ALL
+ default n
+ help
+ Check dependencies between wait and event and report it if
+ deadlock possibility has been detected. Multiple reports are
+ allowed if there are more than a single problem.
+
config LOCK_DEBUGGING_SUPPORT
bool
depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 03/14] dept: Embed Dept data in Lockdep
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 02/14] dept: Implement Dept(Dependency Tracker) Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 04/14] dept: Apply Dept to spinlock Byungchul Park
` (10 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Dept should work independently from Lockdep. However, there's no choise
but to rely on Lockdep code and its instances for now.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 71 ++++++++++++++++++++++++++++++++++++++++---
include/linux/lockdep_types.h | 3 ++
kernel/locking/lockdep.c | 12 ++++----
3 files changed, 76 insertions(+), 10 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 467b942..c56f6b6 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -20,6 +20,33 @@
extern int prove_locking;
extern int lock_stat;
+#ifdef CONFIG_DEPT
+static inline void dept_after_copy_map(struct dept_map *to,
+ struct dept_map *from)
+{
+ int i;
+
+ if (from->keys == &from->keys_local)
+ to->keys = &to->keys_local;
+
+ if (!to->keys)
+ return;
+
+ /*
+ * Since the class cache can be modified concurrently we could observe
+ * half pointers (64bit arch using 32bit copy insns). Therefore clear
+ * the caches and take the performance hit.
+ *
+ * XXX it doesn't work well with lockdep_set_class_and_subclass(), since
+ * that relies on cache abuse.
+ */
+ for (i = 0; i < DEPT_MAX_SUBCLASSES_CACHE; i++)
+ to->keys->classes[i] = NULL;
+}
+#else
+#define dept_after_copy_map(t, f) do { } while (0)
+#endif
+
#ifdef CONFIG_LOCKDEP
#include <linux/linkage.h>
@@ -43,6 +70,8 @@ static inline void lockdep_copy_map(struct lockdep_map *to,
*/
for (i = 0; i < NR_LOCKDEP_CACHING_CLASSES; i++)
to->class_cache[i] = NULL;
+
+ dept_after_copy_map(&to->dmap, &from->dmap);
}
/*
@@ -176,8 +205,19 @@ struct held_lock {
current->lockdep_recursion -= LOCKDEP_OFF; \
} while (0)
-extern void lockdep_register_key(struct lock_class_key *key);
-extern void lockdep_unregister_key(struct lock_class_key *key);
+extern void __lockdep_register_key(struct lock_class_key *key);
+extern void __lockdep_unregister_key(struct lock_class_key *key);
+
+#define lockdep_register_key(k) \
+do { \
+ __lockdep_register_key(k); \
+ dept_key_init(&(k)->dkey); \
+} while (0)
+#define lockdep_unregister_key(k) \
+do { \
+ __lockdep_unregister_key(k); \
+ dept_key_destroy(&(k)->dkey); \
+} while (0)
/*
* These methods are used by specific locking variants (spinlocks,
@@ -185,9 +225,18 @@ struct held_lock {
* to lockdep:
*/
-extern void lockdep_init_map_type(struct lockdep_map *lock, const char *name,
+extern void __lockdep_init_map_type(struct lockdep_map *lock, const char *name,
struct lock_class_key *key, int subclass, u8 inner, u8 outer, u8 lock_type);
+#define lockdep_init_map_type(l, n, k, s, i, o, t) \
+do { \
+ __lockdep_init_map_type(l, n, k, s, i, o, t); \
+ if ((k) == &__lockdep_no_validate__) \
+ dept_map_nocheck(&(l)->dmap); \
+ else \
+ dept_map_init(&(l)->dmap, &(k)->dkey, s, n); \
+} while (0)
+
static inline void
lockdep_init_map_waits(struct lockdep_map *lock, const char *name,
struct lock_class_key *key, int subclass, u8 inner, u8 outer)
@@ -431,13 +480,27 @@ enum xhlock_context_t {
XHLOCK_CTX_NR,
};
+#ifdef CONFIG_DEPT
+/*
+ * TODO: I found the case to use an address of other than a real key as
+ * _key, for instance, in workqueue. So for now, we cannot use the
+ * assignment like '.dmap.keys = &(_key)->dkey' unless it's fixed.
+ */
+#define STATIC_DEPT_MAP_INIT(_name, _key) .dmap = { \
+ .name = (_name), \
+ .keys = NULL },
+#else
+#define STATIC_DEPT_MAP_INIT(_name, _key)
+#endif
+
#define lockdep_init_map_crosslock(m, n, k, s) do {} while (0)
/*
* To initialize a lockdep_map statically use this macro.
* Note that _name must not be NULL.
*/
#define STATIC_LOCKDEP_MAP_INIT(_name, _key) \
- { .name = (_name), .key = (void *)(_key), }
+ { .name = (_name), .key = (void *)(_key), \
+ STATIC_DEPT_MAP_INIT(_name, _key) }
static inline void lockdep_invariant_state(bool force) {}
static inline void lockdep_free_task(struct task_struct *task) {}
diff --git a/include/linux/lockdep_types.h b/include/linux/lockdep_types.h
index d224308..50c8879 100644
--- a/include/linux/lockdep_types.h
+++ b/include/linux/lockdep_types.h
@@ -11,6 +11,7 @@
#define __LINUX_LOCKDEP_TYPES_H
#include <linux/types.h>
+#include <linux/dept.h>
#define MAX_LOCKDEP_SUBCLASSES 8UL
@@ -76,6 +77,7 @@ struct lock_class_key {
struct hlist_node hash_entry;
struct lockdep_subclass_key subkeys[MAX_LOCKDEP_SUBCLASSES];
};
+ struct dept_key dkey;
};
extern struct lock_class_key __lockdep_no_validate__;
@@ -185,6 +187,7 @@ struct lockdep_map {
int cpu;
unsigned long ip;
#endif
+ struct dept_map dmap;
};
struct pin_cookie { unsigned int val; };
diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 4a882f8..a85468d 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -1184,7 +1184,7 @@ static inline struct hlist_head *keyhashentry(const struct lock_class_key *key)
}
/* Register a dynamically allocated key. */
-void lockdep_register_key(struct lock_class_key *key)
+void __lockdep_register_key(struct lock_class_key *key)
{
struct hlist_head *hash_head;
struct lock_class_key *k;
@@ -1207,7 +1207,7 @@ void lockdep_register_key(struct lock_class_key *key)
restore_irqs:
raw_local_irq_restore(flags);
}
-EXPORT_SYMBOL_GPL(lockdep_register_key);
+EXPORT_SYMBOL_GPL(__lockdep_register_key);
/* Check whether a key has been registered as a dynamic key. */
static bool is_dynamic_key(const struct lock_class_key *key)
@@ -4771,7 +4771,7 @@ static inline int check_wait_context(struct task_struct *curr,
/*
* Initialize a lock instance's lock-class mapping info:
*/
-void lockdep_init_map_type(struct lockdep_map *lock, const char *name,
+void __lockdep_init_map_type(struct lockdep_map *lock, const char *name,
struct lock_class_key *key, int subclass,
u8 inner, u8 outer, u8 lock_type)
{
@@ -4831,7 +4831,7 @@ void lockdep_init_map_type(struct lockdep_map *lock, const char *name,
raw_local_irq_restore(flags);
}
}
-EXPORT_SYMBOL_GPL(lockdep_init_map_type);
+EXPORT_SYMBOL_GPL(__lockdep_init_map_type);
struct lock_class_key __lockdep_no_validate__;
EXPORT_SYMBOL_GPL(__lockdep_no_validate__);
@@ -6291,7 +6291,7 @@ void lockdep_reset_lock(struct lockdep_map *lock)
}
/* Unregister a dynamically allocated key. */
-void lockdep_unregister_key(struct lock_class_key *key)
+void __lockdep_unregister_key(struct lock_class_key *key)
{
struct hlist_head *hash_head = keyhashentry(key);
struct lock_class_key *k;
@@ -6326,7 +6326,7 @@ void lockdep_unregister_key(struct lock_class_key *key)
/* Wait until is_dynamic_key() has finished accessing k->hash_entry. */
synchronize_rcu();
}
-EXPORT_SYMBOL_GPL(lockdep_unregister_key);
+EXPORT_SYMBOL_GPL(__lockdep_unregister_key);
void __init lockdep_init(void)
{
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 04/14] dept: Apply Dept to spinlock
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 02/14] dept: Implement Dept(Dependency Tracker) Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 03/14] dept: Embed Dept data in Lockdep Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 05/14] dept: Apply Dept to mutex families Byungchul Park
` (9 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes Dept able to track dependencies by spinlock.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 18 +++++++++++++++---
include/linux/spinlock.h | 24 ++++++++++++++++++++++++
include/linux/spinlock_types_raw.h | 13 +++++++++++++
3 files changed, 52 insertions(+), 3 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index c56f6b6..1da8b95 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -582,9 +582,21 @@ static inline void print_irqtrace_events(struct task_struct *curr)
#define lock_acquire_shared(l, s, t, n, i) lock_acquire(l, s, t, 1, 1, n, i)
#define lock_acquire_shared_recursive(l, s, t, n, i) lock_acquire(l, s, t, 2, 1, n, i)
-#define spin_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
-#define spin_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
-#define spin_release(l, i) lock_release(l, i)
+#define spin_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_spin_lock(&(l)->dmap, s, t, NULL, "spin_unlock", i); \
+} while (0)
+#define spin_acquire_nest(l, s, t, n, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, n, i); \
+ dept_spin_lock(&(l)->dmap, s, t, (n) ? &(n)->dmap : NULL, "spin_unlock", i); \
+} while (0)
+#define spin_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_spin_unlock(&(l)->dmap, i); \
+} while (0)
#define rwlock_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
#define rwlock_acquire_read(l, s, t, i) \
diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index 5c0c517..4d1f5e4 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -95,6 +95,30 @@
# include <linux/spinlock_up.h>
#endif
+#ifdef CONFIG_DEPT
+#define dept_spin_lock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, 1UL, ip, __func__, ne); \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_spin_unlock(m, ip) \
+do { \
+ dept_event(m, 1UL, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_spin_lock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_spin_unlock(m, ip) do { } while (0)
+#endif
+
#ifdef CONFIG_DEBUG_SPINLOCK
extern void __raw_spin_lock_init(raw_spinlock_t *lock, const char *name,
struct lock_class_key *key, short inner);
diff --git a/include/linux/spinlock_types_raw.h b/include/linux/spinlock_types_raw.h
index 91cb36b..5a9b25d 100644
--- a/include/linux/spinlock_types_raw.h
+++ b/include/linux/spinlock_types_raw.h
@@ -26,16 +26,28 @@
#define SPINLOCK_OWNER_INIT ((void *)-1L)
+#ifdef CONFIG_DEPT
+# define RAW_SPIN_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+# define SPIN_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+# define LOCAL_SPIN_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define RAW_SPIN_DMAP_INIT(lockname)
+# define SPIN_DMAP_INIT(lockname)
+# define LOCAL_SPIN_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define RAW_SPIN_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_SPIN, \
+ RAW_SPIN_DMAP_INIT(lockname) \
}
# define SPIN_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_CONFIG, \
+ SPIN_DMAP_INIT(lockname) \
}
# define LOCAL_SPIN_DEP_MAP_INIT(lockname) \
@@ -43,6 +55,7 @@
.name = #lockname, \
.wait_type_inner = LD_WAIT_CONFIG, \
.lock_type = LD_LOCK_PERCPU, \
+ LOCAL_SPIN_DMAP_INIT(lockname) \
}
#else
# define RAW_SPIN_DEP_MAP_INIT(lockname)
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 05/14] dept: Apply Dept to mutex families
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (2 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 04/14] dept: Apply Dept to spinlock Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 06/14] dept: Apply Dept to rwlock Byungchul Park
` (8 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes Dept able to track dependencies by mutex families.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 18 +++++++++++++++---
include/linux/mutex.h | 31 +++++++++++++++++++++++++++++++
include/linux/rtmutex.h | 7 +++++++
3 files changed, 53 insertions(+), 3 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 1da8b95..4c6c2a1 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -613,9 +613,21 @@ static inline void print_irqtrace_events(struct task_struct *curr)
#define seqcount_acquire_read(l, s, t, i) lock_acquire_shared_recursive(l, s, t, NULL, i)
#define seqcount_release(l, i) lock_release(l, i)
-#define mutex_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
-#define mutex_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
-#define mutex_release(l, i) lock_release(l, i)
+#define mutex_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_mutex_lock(&(l)->dmap, s, t, NULL, "mutex_unlock", i); \
+} while (0)
+#define mutex_acquire_nest(l, s, t, n, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, n, i); \
+ dept_mutex_lock(&(l)->dmap, s, t, (n) ? &(n)->dmap : NULL, "mutex_unlock", i);\
+} while (0)
+#define mutex_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_mutex_unlock(&(l)->dmap, i); \
+} while (0)
#define rwsem_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
#define rwsem_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
diff --git a/include/linux/mutex.h b/include/linux/mutex.h
index 8f226d4..0955760 100644
--- a/include/linux/mutex.h
+++ b/include/linux/mutex.h
@@ -20,11 +20,18 @@
#include <linux/osq_lock.h>
#include <linux/debug_locks.h>
+#ifdef CONFIG_DEPT
+# define DMAP_MUTEX_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define DMAP_MUTEX_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define __DEP_MAP_MUTEX_INITIALIZER(lockname) \
, .dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_SLEEP, \
+ DMAP_MUTEX_INIT(lockname) \
}
#else
# define __DEP_MAP_MUTEX_INITIALIZER(lockname)
@@ -75,6 +82,30 @@ struct mutex {
#endif
};
+#ifdef CONFIG_DEPT
+#define dept_mutex_lock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, 1UL, ip, __func__, ne); \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_mutex_unlock(m, ip) \
+do { \
+ dept_event(m, 1UL, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_mutex_lock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_mutex_unlock(m, ip) do { } while (0)
+#endif
+
#ifdef CONFIG_DEBUG_MUTEXES
#define __DEBUG_MUTEX_INITIALIZER(lockname) \
diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h
index 7d04988..60cebb0 100644
--- a/include/linux/rtmutex.h
+++ b/include/linux/rtmutex.h
@@ -76,11 +76,18 @@ static inline void rt_mutex_debug_task_free(struct task_struct *tsk) { }
__rt_mutex_init(mutex, __func__, &__key); \
} while (0)
+#ifdef CONFIG_DEPT
+#define DMAP_RT_MUTEX_INIT(mutexname) .dmap = { .name = #mutexname },
+#else
+#define DMAP_RT_MUTEX_INIT(mutexname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
#define __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname) \
.dep_map = { \
.name = #mutexname, \
.wait_type_inner = LD_WAIT_SLEEP, \
+ DMAP_RT_MUTEX_INIT(mutexname) \
}
#else
#define __DEP_MAP_RT_MUTEX_INITIALIZER(mutexname)
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 06/14] dept: Apply Dept to rwlock
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (3 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 05/14] dept: Apply Dept to mutex families Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 07/14] dept: Apply Dept to wait_for_completion()/complete() Byungchul Park
` (7 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes Dept able to track dependencies by rwlock.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 25 +++++++++++++++++-----
include/linux/rwlock.h | 48 ++++++++++++++++++++++++++++++++++++++++++
include/linux/rwlock_api_smp.h | 8 +++----
include/linux/rwlock_types.h | 7 ++++++
4 files changed, 79 insertions(+), 9 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 4c6c2a1..306c22d 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -598,16 +598,31 @@ static inline void print_irqtrace_events(struct task_struct *curr)
dept_spin_unlock(&(l)->dmap, i); \
} while (0)
-#define rwlock_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
+#define rwlock_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_rwlock_wlock(&(l)->dmap, s, t, NULL, "write_unlock", i); \
+} while (0)
#define rwlock_acquire_read(l, s, t, i) \
do { \
- if (read_lock_is_recursive()) \
+ if (read_lock_is_recursive()) { \
lock_acquire_shared_recursive(l, s, t, NULL, i); \
- else \
+ dept_rwlock_rlock(&(l)->dmap, s, t, NULL, "read_unlock", i, 0);\
+ } else { \
lock_acquire_shared(l, s, t, NULL, i); \
+ dept_rwlock_rlock(&(l)->dmap, s, t, NULL, "read_unlock", i, 1);\
+ } \
+} while (0)
+#define rwlock_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_rwlock_wunlock(&(l)->dmap, i); \
+} while (0)
+#define rwlock_release_read(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_rwlock_runlock(&(l)->dmap, i); \
} while (0)
-
-#define rwlock_release(l, i) lock_release(l, i)
#define seqcount_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
#define seqcount_acquire_read(l, s, t, i) lock_acquire_shared_recursive(l, s, t, NULL, i)
diff --git a/include/linux/rwlock.h b/include/linux/rwlock.h
index 8f416c5..7bc8c40 100644
--- a/include/linux/rwlock.h
+++ b/include/linux/rwlock.h
@@ -28,6 +28,54 @@
do { *(lock) = __RW_LOCK_UNLOCKED(lock); } while (0)
#endif
+#ifdef CONFIG_DEPT
+#define DEPT_EVT_RWLOCK_R 1UL
+#define DEPT_EVT_RWLOCK_W (1UL << 1)
+#define DEPT_EVT_RWLOCK_RW (DEPT_EVT_RWLOCK_R | DEPT_EVT_RWLOCK_W)
+
+#define dept_rwlock_wlock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_W, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, DEPT_EVT_RWLOCK_RW, ip, __func__, ne); \
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_W, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } \
+}while (0)
+#define dept_rwlock_rlock(m, ne, t, n, e_fn, ip, q) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_R, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, (q) ? DEPT_EVT_RWLOCK_RW : DEPT_EVT_RWLOCK_W, ip, __func__, ne);\
+ dept_ecxt_enter(m, DEPT_EVT_RWLOCK_R, ip, __func__, e_fn, ne);\
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_rwlock_wunlock(m, ip) \
+do { \
+ dept_event(m, DEPT_EVT_RWLOCK_W, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#define dept_rwlock_runlock(m, ip) \
+do { \
+ dept_event(m, DEPT_EVT_RWLOCK_R, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_rwlock_wlock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_rwlock_rlock(m, ne, t, n, e_fn, ip, q) do { } while (0)
+#define dept_rwlock_wunlock(m, ip) do { } while (0)
+#define dept_rwlock_runlock(m, ip) do { } while (0)
+#endif
+
#ifdef CONFIG_DEBUG_SPINLOCK
extern void do_raw_read_lock(rwlock_t *lock) __acquires(lock);
extern int do_raw_read_trylock(rwlock_t *lock);
diff --git a/include/linux/rwlock_api_smp.h b/include/linux/rwlock_api_smp.h
index dceb0a5..a222cf1 100644
--- a/include/linux/rwlock_api_smp.h
+++ b/include/linux/rwlock_api_smp.h
@@ -228,7 +228,7 @@ static inline void __raw_write_unlock(rwlock_t *lock)
static inline void __raw_read_unlock(rwlock_t *lock)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
preempt_enable();
}
@@ -236,7 +236,7 @@ static inline void __raw_read_unlock(rwlock_t *lock)
static inline void
__raw_read_unlock_irqrestore(rwlock_t *lock, unsigned long flags)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
local_irq_restore(flags);
preempt_enable();
@@ -244,7 +244,7 @@ static inline void __raw_read_unlock(rwlock_t *lock)
static inline void __raw_read_unlock_irq(rwlock_t *lock)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
local_irq_enable();
preempt_enable();
@@ -252,7 +252,7 @@ static inline void __raw_read_unlock_irq(rwlock_t *lock)
static inline void __raw_read_unlock_bh(rwlock_t *lock)
{
- rwlock_release(&lock->dep_map, _RET_IP_);
+ rwlock_release_read(&lock->dep_map, _RET_IP_);
do_raw_read_unlock(lock);
__local_bh_enable_ip(_RET_IP_, SOFTIRQ_LOCK_OFFSET);
}
diff --git a/include/linux/rwlock_types.h b/include/linux/rwlock_types.h
index 1948442..6dddc5b 100644
--- a/include/linux/rwlock_types.h
+++ b/include/linux/rwlock_types.h
@@ -5,11 +5,18 @@
# error "Do not include directly, include spinlock_types.h"
#endif
+#ifdef CONFIG_DEPT
+# define RW_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define RW_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define RW_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_CONFIG, \
+ RW_DMAP_INIT(lockname) \
}
#else
# define RW_DEP_MAP_INIT(lockname)
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 07/14] dept: Apply Dept to wait_for_completion()/complete()
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (4 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 06/14] dept: Apply Dept to rwlock Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 08/14] dept: Apply Dept to seqlock Byungchul Park
` (6 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes Dept able to track dependencies by
wait_for_completion()/complete().
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/completion.h | 42 ++++++++++++++++++++++++++++++++++++++++--
kernel/sched/completion.c | 12 ++++++++++--
2 files changed, 50 insertions(+), 4 deletions(-)
diff --git a/include/linux/completion.h b/include/linux/completion.h
index 51d9ab0..2bba7b1 100644
--- a/include/linux/completion.h
+++ b/include/linux/completion.h
@@ -26,14 +26,48 @@
struct completion {
unsigned int done;
struct swait_queue_head wait;
+ struct dept_map dmap;
};
+#ifdef CONFIG_DEPT
+#define dept_wfc_init(m, k, s, n) dept_map_init(m, k, s, n)
+#define dept_wfc_reinit(m) dept_map_reinit(m)
+#define dept_wfc_wait(m, ip) \
+do { \
+ dept_asked_event(m); \
+ dept_wait(m, 1UL, ip, __func__, 0); \
+} while (0)
+#define dept_wfc_complete(m, ip) dept_event(m, 1UL, ip, __func__)
+#define dept_wfc_enter(m, ip) dept_ecxt_enter(m, 1UL, ip, "completion_context_enter", "complete", 0)
+#define dept_wfc_exit(m, ip) dept_ecxt_exit(m, ip)
+#else
+#define dept_wfc_init(m, k, s, n) do { (void)(n); (void)(k); } while (0)
+#define dept_wfc_reinit(m) do { } while (0)
+#define dept_wfc_wait(m, ip) do { } while (0)
+#define dept_wfc_complete(m, ip) do { } while (0)
+#define dept_wfc_enter(m, ip) do { } while (0)
+#define dept_wfc_exit(m, ip) do { } while (0)
+#endif
+
+#ifdef CONFIG_DEPT
+#define WFC_DEPT_MAP_INIT(work) .dmap = { .name = #work }
+#else
+#define WFC_DEPT_MAP_INIT(work)
+#endif
+
+#define init_completion(x) \
+ do { \
+ static struct dept_key __dkey; \
+ __init_completion(x, &__dkey, #x); \
+ } while (0)
+
#define init_completion_map(x, m) init_completion(x)
static inline void complete_acquire(struct completion *x) {}
static inline void complete_release(struct completion *x) {}
#define COMPLETION_INITIALIZER(work) \
- { 0, __SWAIT_QUEUE_HEAD_INITIALIZER((work).wait) }
+ { 0, __SWAIT_QUEUE_HEAD_INITIALIZER((work).wait), \
+ WFC_DEPT_MAP_INIT(work) }
#define COMPLETION_INITIALIZER_ONSTACK_MAP(work, map) \
(*({ init_completion_map(&(work), &(map)); &(work); }))
@@ -81,9 +115,12 @@ static inline void complete_release(struct completion *x) {}
* This inline function will initialize a dynamically created completion
* structure.
*/
-static inline void init_completion(struct completion *x)
+static inline void __init_completion(struct completion *x,
+ struct dept_key *dkey,
+ const char *name)
{
x->done = 0;
+ dept_wfc_init(&x->dmap, dkey, 0, name);
init_swait_queue_head(&x->wait);
}
@@ -97,6 +134,7 @@ static inline void init_completion(struct completion *x)
static inline void reinit_completion(struct completion *x)
{
x->done = 0;
+ dept_wfc_reinit(&x->dmap);
}
extern void wait_for_completion(struct completion *);
diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c
index a778554..6e31cc0 100644
--- a/kernel/sched/completion.c
+++ b/kernel/sched/completion.c
@@ -29,6 +29,7 @@ void complete(struct completion *x)
{
unsigned long flags;
+ dept_wfc_complete(&x->dmap, _RET_IP_);
raw_spin_lock_irqsave(&x->wait.lock, flags);
if (x->done != UINT_MAX)
@@ -58,6 +59,7 @@ void complete_all(struct completion *x)
{
unsigned long flags;
+ dept_wfc_complete(&x->dmap, _RET_IP_);
lockdep_assert_RT_in_threaded_ctx();
raw_spin_lock_irqsave(&x->wait.lock, flags);
@@ -112,17 +114,23 @@ void complete_all(struct completion *x)
}
static long __sched
-wait_for_common(struct completion *x, long timeout, int state)
+_wait_for_common(struct completion *x, long timeout, int state)
{
return __wait_for_common(x, schedule_timeout, timeout, state);
}
static long __sched
-wait_for_common_io(struct completion *x, long timeout, int state)
+_wait_for_common_io(struct completion *x, long timeout, int state)
{
return __wait_for_common(x, io_schedule_timeout, timeout, state);
}
+#define wait_for_common(x, t, s) \
+({ dept_wfc_wait(&(x)->dmap, _RET_IP_); _wait_for_common(x, t, s); })
+
+#define wait_for_common_io(x, t, s) \
+({ dept_wfc_wait(&(x)->dmap, _RET_IP_); _wait_for_common_io(x, t, s); })
+
/**
* wait_for_completion: - waits for completion of a task
* @x: holds the state of this particular completion
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 08/14] dept: Apply Dept to seqlock
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (5 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 07/14] dept: Apply Dept to wait_for_completion()/complete() Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 09/14] dept: Apply Dept to rwsem Byungchul Park
` (5 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes Dept able to track dependencies by seqlock with adding wait
annotation on read side of seqlock.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/seqlock.h | 19 ++++++++++++++++++-
1 file changed, 18 insertions(+), 1 deletion(-)
diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h
index 37ded6b..35929c9 100644
--- a/include/linux/seqlock.h
+++ b/include/linux/seqlock.h
@@ -23,6 +23,13 @@
#include <asm/processor.h>
+#ifdef CONFIG_DEPT
+#define DEPT_EVT_ALL ((1UL << DEPT_MAX_SUBCLASSES_EVT) - 1)
+#define dept_seq_wait(m, ip) dept_wait(m, DEPT_EVT_ALL, ip, __func__, 0)
+#else
+#define dept_seq_wait(m, ip) do { } while (0)
+#endif
+
/*
* The seqlock seqcount_t interface does not prescribe a precise sequence of
* read begin/retry/end. For readers, typically there is a call to
@@ -148,7 +155,7 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s)
* This lock-unlock technique must be implemented for all of PREEMPT_RT
* sleeping locks. See Documentation/locking/locktypes.rst
*/
-#if defined(CONFIG_LOCKDEP) || defined(CONFIG_PREEMPT_RT)
+#if defined(CONFIG_LOCKDEP) || defined(CONFIG_DEPT) || defined(CONFIG_PREEMPT_RT)
#define __SEQ_LOCK(expr) expr
#else
#define __SEQ_LOCK(expr)
@@ -203,6 +210,12 @@ static inline void seqcount_lockdep_reader_access(const seqcount_t *s)
__SEQ_LOCK(locktype *lock); \
} seqcount_##lockname##_t; \
\
+static __always_inline void \
+__seqprop_##lockname##_wait(const seqcount_##lockname##_t *s) \
+{ \
+ __SEQ_LOCK(dept_seq_wait(&(lockmember)->dep_map.dmap, _RET_IP_));\
+} \
+ \
static __always_inline seqcount_t * \
__seqprop_##lockname##_ptr(seqcount_##lockname##_t *s) \
{ \
@@ -271,6 +284,8 @@ static inline void __seqprop_assert(const seqcount_t *s)
lockdep_assert_preemption_disabled();
}
+static inline void __seqprop_wait(const seqcount_t *s) { }
+
#define __SEQ_RT IS_ENABLED(CONFIG_PREEMPT_RT)
SEQCOUNT_LOCKNAME(raw_spinlock, raw_spinlock_t, false, s->lock, raw_spin, raw_spin_lock(s->lock))
@@ -311,6 +326,7 @@ static inline void __seqprop_assert(const seqcount_t *s)
#define seqprop_sequence(s) __seqprop(s, sequence)
#define seqprop_preemptible(s) __seqprop(s, preemptible)
#define seqprop_assert(s) __seqprop(s, assert)
+#define seqprop_dept_wait(s) __seqprop(s, wait)
/**
* __read_seqcount_begin() - begin a seqcount_t read section w/o barrier
@@ -360,6 +376,7 @@ static inline void __seqprop_assert(const seqcount_t *s)
#define read_seqcount_begin(s) \
({ \
seqcount_lockdep_reader_access(seqprop_ptr(s)); \
+ seqprop_dept_wait(s); \
raw_read_seqcount_begin(s); \
})
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 09/14] dept: Apply Dept to rwsem
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (6 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 08/14] dept: Apply Dept to seqlock Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 10/14] dept: Add proc knobs to show stats and dependency graph Byungchul Park
` (4 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes Dept able to track dependencies by rwsem.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/lockdep.h | 24 ++++++++++++++++++++----
include/linux/percpu-rwsem.h | 10 +++++++++-
include/linux/rwsem.h | 31 +++++++++++++++++++++++++++++++
3 files changed, 60 insertions(+), 5 deletions(-)
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 306c22d..6aab26c 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -644,10 +644,26 @@ static inline void print_irqtrace_events(struct task_struct *curr)
dept_mutex_unlock(&(l)->dmap, i); \
} while (0)
-#define rwsem_acquire(l, s, t, i) lock_acquire_exclusive(l, s, t, NULL, i)
-#define rwsem_acquire_nest(l, s, t, n, i) lock_acquire_exclusive(l, s, t, n, i)
-#define rwsem_acquire_read(l, s, t, i) lock_acquire_shared(l, s, t, NULL, i)
-#define rwsem_release(l, i) lock_release(l, i)
+#define rwsem_acquire(l, s, t, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, NULL, i); \
+ dept_rwsem_lock(&(l)->dmap, s, t, NULL, "up_write", i); \
+} while (0)
+#define rwsem_acquire_nest(l, s, t, n, i) \
+do { \
+ lock_acquire_exclusive(l, s, t, n, i); \
+ dept_rwsem_lock(&(l)->dmap, s, t, (n) ? &(n)->dmap : NULL, "up_write", i);\
+} while (0)
+#define rwsem_acquire_read(l, s, t, i) \
+do { \
+ lock_acquire_shared(l, s, t, NULL, i); \
+ dept_rwsem_lock(&(l)->dmap, s, t, NULL, "up_read", i); \
+} while (0)
+#define rwsem_release(l, i) \
+do { \
+ lock_release(l, i); \
+ dept_rwsem_unlock(&(l)->dmap, i); \
+} while (0)
#define lock_map_acquire(l) lock_acquire_exclusive(l, 0, 0, NULL, _THIS_IP_)
#define lock_map_acquire_read(l) lock_acquire_shared_recursive(l, 0, 0, NULL, _THIS_IP_)
diff --git a/include/linux/percpu-rwsem.h b/include/linux/percpu-rwsem.h
index 5fda40f..7ec5625 100644
--- a/include/linux/percpu-rwsem.h
+++ b/include/linux/percpu-rwsem.h
@@ -20,8 +20,16 @@ struct percpu_rw_semaphore {
#endif
};
+#ifdef CONFIG_DEPT
+#define __PERCPU_RWSEM_DMAP_INIT(lockname) .dmap = { .name = #lockname }
+#else
+#define __PERCPU_RWSEM_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
-#define __PERCPU_RWSEM_DEP_MAP_INIT(lockname) .dep_map = { .name = #lockname },
+#define __PERCPU_RWSEM_DEP_MAP_INIT(lockname) .dep_map = { \
+ .name = #lockname, \
+ __PERCPU_RWSEM_DMAP_INIT(lockname) },
#else
#define __PERCPU_RWSEM_DEP_MAP_INIT(lockname)
#endif
diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h
index f934876..ec3bc1d 100644
--- a/include/linux/rwsem.h
+++ b/include/linux/rwsem.h
@@ -16,11 +16,18 @@
#include <linux/atomic.h>
#include <linux/err.h>
+#ifdef CONFIG_DEPT
+# define RWSEM_DMAP_INIT(lockname) .dmap = { .name = #lockname },
+#else
+# define RWSEM_DMAP_INIT(lockname)
+#endif
+
#ifdef CONFIG_DEBUG_LOCK_ALLOC
# define __RWSEM_DEP_MAP_INIT(lockname) \
.dep_map = { \
.name = #lockname, \
.wait_type_inner = LD_WAIT_SLEEP, \
+ RWSEM_DMAP_INIT(lockname) \
},
#else
# define __RWSEM_DEP_MAP_INIT(lockname)
@@ -32,6 +39,30 @@
#include <linux/osq_lock.h>
#endif
+#ifdef CONFIG_DEPT
+#define dept_rwsem_lock(m, ne, t, n, e_fn, ip) \
+do { \
+ if (t) { \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } else if (n) { \
+ dept_warn_on(dept_top_map() != (n)); \
+ } else { \
+ dept_wait(m, 1UL, ip, __func__, ne); \
+ dept_ecxt_enter(m, 1UL, ip, __func__, e_fn, ne); \
+ dept_asked_event(m); \
+ } \
+} while (0)
+#define dept_rwsem_unlock(m, ip) \
+do { \
+ dept_event(m, 1UL, ip, __func__); \
+ dept_ecxt_exit(m, ip); \
+} while (0)
+#else
+#define dept_rwsem_lock(m, ne, t, n, e_fn, ip) do { } while (0)
+#define dept_rwsem_unlock(m, ip) do { } while (0)
+#endif
+
/*
* For an uncontended rwsem, count and owner are the only fields a task
* needs to touch when acquiring the rwsem. So they are put next to each
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 10/14] dept: Add proc knobs to show stats and dependency graph
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (7 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 09/14] dept: Apply Dept to rwsem Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 11/14] dept: Introduce split map concept and new APIs for them Byungchul Park
` (3 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
It'd be useful to show Dept internal stats and dependency graph on
runtime via proc for better information. Introduced the knobs.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
kernel/dependency/Makefile | 1 +
kernel/dependency/dept.c | 24 ++++------
kernel/dependency/dept_internal.h | 26 +++++++++++
kernel/dependency/dept_proc.c | 97 +++++++++++++++++++++++++++++++++++++++
4 files changed, 133 insertions(+), 15 deletions(-)
create mode 100644 kernel/dependency/dept_internal.h
create mode 100644 kernel/dependency/dept_proc.c
diff --git a/kernel/dependency/Makefile b/kernel/dependency/Makefile
index 9f7778e..49152ad 100644
--- a/kernel/dependency/Makefile
+++ b/kernel/dependency/Makefile
@@ -1,4 +1,5 @@
# SPDX-License-Identifier: GPL-2.0
obj-$(CONFIG_DEPT) += dept.o
+obj-$(CONFIG_DEPT) += dept_proc.o
diff --git a/kernel/dependency/dept.c b/kernel/dependency/dept.c
index 64287894..b682bf9 100644
--- a/kernel/dependency/dept.c
+++ b/kernel/dependency/dept.c
@@ -90,6 +90,7 @@
#include <linux/hash.h>
#include <linux/dept.h>
#include <linux/utsname.h>
+#include "dept_internal.h"
static int dept_stop;
static int dept_per_cpu_ready;
@@ -241,20 +242,13 @@ static inline struct dept_task *dept_task(void)
* have been freed will be placed.
*/
-enum object_t {
-#define OBJECT(id, nr) OBJECT_##id,
- #include "dept_object.h"
-#undef OBJECT
- OBJECT_NR,
-};
-
#define OBJECT(id, nr) \
static struct dept_##id spool_##id[nr]; \
static DEFINE_PER_CPU(struct llist_head, lpool_##id);
#include "dept_object.h"
#undef OBJECT
-static struct dept_pool pool[OBJECT_NR] = {
+struct dept_pool dept_pool[OBJECT_NR] = {
#define OBJECT(id, nr) { \
.name = #id, \
.obj_sz = sizeof(struct dept_##id), \
@@ -283,7 +277,7 @@ static void *from_pool(enum object_t t)
if (DEPT_WARN_ON(!irqs_disabled()))
return NULL;
- p = &pool[t];
+ p = &dept_pool[t];
/*
* Try local pool first.
@@ -312,7 +306,7 @@ static void *from_pool(enum object_t t)
static void to_pool(void *o, enum object_t t)
{
- struct dept_pool *p = &pool[t];
+ struct dept_pool *p = &dept_pool[t];
struct llist_head *h;
preempt_disable();
@@ -1986,7 +1980,7 @@ void dept_map_nocheck(struct dept_map *m)
}
EXPORT_SYMBOL_GPL(dept_map_nocheck);
-static LIST_HEAD(classes);
+LIST_HEAD(dept_classes);
static inline bool within(const void *addr, void *start, unsigned long size)
{
@@ -2031,7 +2025,7 @@ void dept_free_range(void *start, unsigned int sz)
*/
free_staleie_staleiw_range(start, sz);
- list_for_each_entry_safe(c, n, &classes, all_node) {
+ list_for_each_entry_safe(c, n, &dept_classes, all_node) {
if (!within((void *)c->key, start, sz) &&
!within(c->name, start, sz))
continue;
@@ -2100,7 +2094,7 @@ static struct dept_class *check_new_class(struct dept_key *local,
c->sub = sub;
c->key = (unsigned long)(k->subkeys + sub);
hash_add_class(c);
- list_add(&c->all_node, &classes);
+ list_add(&c->all_node, &dept_classes);
unlock:
dept_unlock();
caching:
@@ -2429,8 +2423,8 @@ static void migrate_per_cpu_pool(void)
struct llist_head *from;
struct llist_head *to;
- from = &pool[i].boot_pool;
- to = per_cpu_ptr(pool[i].lpool, boot_cpu);
+ from = &dept_pool[i].boot_pool;
+ to = per_cpu_ptr(dept_pool[i].lpool, boot_cpu);
move_llist(to, from);
}
}
diff --git a/kernel/dependency/dept_internal.h b/kernel/dependency/dept_internal.h
new file mode 100644
index 0000000..007c1ee
--- /dev/null
+++ b/kernel/dependency/dept_internal.h
@@ -0,0 +1,26 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept(DEPendency Tracker) - runtime dependency tracker internal header
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ */
+
+#ifndef __DEPT_INTERNAL_H
+#define __DEPT_INTERNAL_H
+
+#ifdef CONFIG_DEPT
+
+enum object_t {
+#define OBJECT(id, nr) OBJECT_##id,
+ #include "dept_object.h"
+#undef OBJECT
+ OBJECT_NR,
+};
+
+extern struct list_head dept_classes;
+extern struct dept_pool dept_pool[];
+
+#endif
+#endif /* __DEPT_INTERNAL_H */
diff --git a/kernel/dependency/dept_proc.c b/kernel/dependency/dept_proc.c
new file mode 100644
index 0000000..5e5afa3
--- /dev/null
+++ b/kernel/dependency/dept_proc.c
@@ -0,0 +1,97 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * kernel/dependency/dept_proc.c
+ *
+ * Runtime dependency tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (C) 2021 LG Electronics, Inc. , Byungchul Park
+ *
+ * Code for /proc/dept_*:
+ *
+ */
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/dept.h>
+#include "dept_internal.h"
+
+static void *l_next(struct seq_file *m, void *v, loff_t *pos)
+{
+ /*
+ * XXX: Serialize list traversal if needed. The following might
+ * give a wrong information on contention.
+ */
+ return seq_list_next(v, &dept_classes, pos);
+}
+
+static void *l_start(struct seq_file *m, loff_t *pos)
+{
+ /*
+ * XXX: Serialize list traversal if needed. The following might
+ * give a wrong information on contention.
+ */
+ return seq_list_start_head(&dept_classes, *pos);
+}
+
+static void l_stop(struct seq_file *m, void *v)
+{
+}
+
+static int l_show(struct seq_file *m, void *v)
+{
+ struct dept_class *fc = list_entry(v, struct dept_class, all_node);
+ struct dept_dep *d;
+
+ if (v == &dept_classes) {
+ seq_printf(m, "All classes:\n\n");
+ return 0;
+ }
+
+ seq_printf(m, "[%p] %s\n", (void *)fc->key, fc->name);
+
+ /*
+ * XXX: Serialize list traversal if needed. The following might
+ * give a wrong information on contention.
+ */
+ list_for_each_entry(d, &fc->dep_head, dep_node) {
+ struct dept_class *tc = d->wait->class;
+ seq_printf(m, " -> [%p] %s\n", (void *)tc->key, tc->name);
+ }
+ seq_puts(m, "\n");
+
+ return 0;
+}
+
+static const struct seq_operations dept_deps_ops = {
+ .start = l_start,
+ .next = l_next,
+ .stop = l_stop,
+ .show = l_show,
+};
+
+static int dept_stats_show(struct seq_file *m, void *v)
+{
+ int r;
+
+ seq_printf(m, "Availablity in the static pools:\n\n");
+#define OBJECT(id, nr) \
+ r = atomic_read(&dept_pool[OBJECT_##id].obj_nr); \
+ if (r < 0) \
+ r = 0; \
+ seq_printf(m, "%s\t%d/%d(%d%%)\n", #id, r, nr, (r * 100) / (nr));
+ #include "dept_object.h"
+#undef OBJECT
+
+ return 0;
+}
+
+static int __init dept_proc_init(void)
+{
+ proc_create_seq("dept_deps", S_IRUSR, NULL, &dept_deps_ops);
+ proc_create_single("dept_stats", S_IRUSR, NULL, dept_stats_show);
+ return 0;
+}
+
+__initcall(dept_proc_init);
+
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 11/14] dept: Introduce split map concept and new APIs for them
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (8 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 10/14] dept: Add proc knobs to show stats and dependency graph Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback} Byungchul Park
` (2 subsequent siblings)
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
There is a case where all maps used for a type of wait/event is so large
in size. For instance, struct page can be a type for (un)lock_page().
The additional memory size for the maps would be 'the # of pages *
sizeof(struct dept_map)' if each struct page keeps its map all the way,
which might be too big to accept in some system.
It'd better to have splitted map, one is for each instance and the other
is for the type which is commonly used, and new APIs using them. So
introduced split map and new APIs for them.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept.h | 36 ++++++++++++++
kernel/dependency/dept.c | 122 +++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 158 insertions(+)
diff --git a/include/linux/dept.h b/include/linux/dept.h
index 8665629..5a726bf 100644
--- a/include/linux/dept.h
+++ b/include/linux/dept.h
@@ -441,6 +441,30 @@ struct dept_map {
bool nocheck;
};
+struct dept_map_each {
+ /*
+ * wait timestamp associated to this map
+ */
+ unsigned int wgen;
+};
+
+struct dept_map_common {
+ const char *name;
+ struct dept_key *keys;
+ int sub_usr;
+
+ /*
+ * It's local copy for fast acces to the associated classes. And
+ * Also used for dept_key instance for statically defined map.
+ */
+ struct dept_key keys_local;
+
+ /*
+ * whether this map should be going to be checked or not
+ */
+ bool nocheck;
+};
+
#define DEPT_TASK_INITIALIZER(t) \
.dept_task.wait_hist = { { .wait = NULL, } }, \
.dept_task.ecxt_held_pos = 0, \
@@ -469,6 +493,11 @@ struct dept_map {
extern void dept_ecxt_exit(struct dept_map *m, unsigned long ip);
extern struct dept_map *dept_top_map(void);
extern void dept_warn_on(bool cond);
+extern void dept_split_map_each_init(struct dept_map_each *me);
+extern void dept_split_map_common_init(struct dept_map_common *mc, struct dept_key *k, const char *n);
+extern void dept_wait_split_map(struct dept_map_each *me, struct dept_map_common *mc, unsigned long ip, const char *w_fn, int ne);
+extern void dept_event_split_map(struct dept_map_each *me, struct dept_map_common *mc, unsigned long ip, const char *e_fn);
+extern void dept_asked_event_split_map(struct dept_map_each *me, struct dept_map_common *mc);
/*
* for users who want to manage external keys
@@ -505,6 +534,8 @@ struct dept_map {
struct dept_task { };
struct dept_key { };
struct dept_map { };
+struct dept_map_each { };
+struct dept_map_commmon { };
#define DEPT_TASK_INITIALIZER(t)
@@ -525,6 +556,11 @@ struct dept_map {
#define dept_ecxt_exit(m, ip) do { } while (0)
#define dept_top_map() NULL
#define dept_warn_on(c) do { } while (0)
+#define dept_split_map_each_init(me) do { } while (0)
+#define dept_split_map_common_init(mc, k, n) do { (void)(n); (void)(k); } while (0)
+#define dept_wait_split_map(me, mc, ip, w_fn, ne) do { } while (0)
+#define dept_event_split_map(me, mc, ip, e_fn) do { } while (0)
+#define dept_asked_event_split_map(me, mc) do { } while (0)
#define dept_key_init(k) do { (void)(k); } while (0)
#define dept_key_destroy(k) do { (void)(k); } while (0)
diff --git a/kernel/dependency/dept.c b/kernel/dependency/dept.c
index b682bf9..bc764cc 100644
--- a/kernel/dependency/dept.c
+++ b/kernel/dependency/dept.c
@@ -2233,6 +2233,128 @@ void dept_event(struct dept_map *m, unsigned long e_f, unsigned long ip,
}
EXPORT_SYMBOL_GPL(dept_event);
+void dept_split_map_each_init(struct dept_map_each *me)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ me->wgen = 0U;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_split_map_each_init);
+
+void dept_split_map_common_init(struct dept_map_common *mc,
+ struct dept_key *k, const char *n)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ flags = dept_enter();
+
+ if (mc->keys != k)
+ mc->keys = k;
+ clean_classes_cache(&mc->keys_local);
+
+ /*
+ * sub_usr is not used with split map.
+ */
+ mc->sub_usr = 0;
+ mc->name = n;
+ mc->nocheck = false;
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_split_map_common_init);
+
+void dept_wait_split_map(struct dept_map_each *me,
+ struct dept_map_common *mc,
+ unsigned long ip, const char *w_fn, int ne)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_class *c;
+ struct dept_key *k;
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (mc->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ k = mc->keys ?: &mc->keys_local;
+ c = check_new_class(&mc->keys_local, k, 0, mc->name);
+ if (c)
+ add_wait(c, ip, w_fn, ne);
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_wait_split_map);
+
+void dept_asked_event_split_map(struct dept_map_each *me,
+ struct dept_map_common *mc)
+{
+ struct dept_task *dt = dept_task();
+ unsigned long flags;
+ unsigned int wg;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (mc->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ /*
+ * Avoid zero wgen.
+ */
+ wg = atomic_inc_return(&wgen) ?: atomic_inc_return(&wgen);
+ WRITE_ONCE(me->wgen, wg);
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_asked_event_split_map);
+
+void dept_event_split_map(struct dept_map_each *me,
+ struct dept_map_common *mc,
+ unsigned long ip, const char *e_fn)
+{
+ struct dept_task *dt = dept_task();
+ struct dept_class *c;
+ struct dept_key *k;
+ unsigned long flags;
+
+ if (READ_ONCE(dept_stop) || dt->recursive)
+ return;
+
+ if (mc->nocheck)
+ return;
+
+ flags = dept_enter();
+
+ k = mc->keys ?: &mc->keys_local;
+ c = check_new_class(&mc->keys_local, k, 0, mc->name);
+ if (c) {
+ add_ecxt((void *)me, c, 0UL, NULL, e_fn, 0);
+ do_event((void *)me, c, READ_ONCE(me->wgen), ip);
+ pop_ecxt((void *)me);
+ }
+
+ dept_exit(flags);
+}
+EXPORT_SYMBOL_GPL(dept_event_split_map);
+
void dept_ecxt_exit(struct dept_map *m, unsigned long ip)
{
struct dept_task *dt = dept_task();
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback}
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (9 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 11/14] dept: Introduce split map concept and new APIs for them Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 14/14] dept: Apply SDT to swait Byungchul Park
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes Dept able to track dependencies by PG_{locked,writeback}. For
instance, (un)lock_page() generates that type of dependency.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept_page.h | 71 +++++++++++++++++++++++++++++++++++++++++
include/linux/page-flags.h | 43 +++++++++++++++++++++++--
include/linux/pagemap.h | 6 +++-
init/main.c | 2 ++
kernel/dependency/dept_object.h | 2 +-
lib/Kconfig.debug | 1 +
mm/filemap.c | 62 +++++++++++++++++++++++++++++++++++
mm/page_ext.c | 5 +++
8 files changed, 188 insertions(+), 4 deletions(-)
create mode 100644 include/linux/dept_page.h
diff --git a/include/linux/dept_page.h b/include/linux/dept_page.h
new file mode 100644
index 0000000..d5a79af
--- /dev/null
+++ b/include/linux/dept_page.h
@@ -0,0 +1,71 @@
+#ifndef __LINUX_DEPT_PAGE_H
+#define __LINUX_DEPT_PAGE_H
+
+#ifdef CONFIG_DEPT
+#include <linux/dept.h>
+
+extern struct page_ext_operations dept_pglocked_ops;
+extern struct page_ext_operations dept_pgwriteback_ops;
+extern struct dept_map_common pglocked_mc;
+extern struct dept_map_common pgwriteback_mc;
+
+extern void dept_page_init(void);
+extern struct dept_map_each *get_pglocked_me(struct page *page);
+extern struct dept_map_each *get_pgwriteback_me(struct page *page);
+
+#define dept_pglocked_wait(f) \
+do { \
+ struct dept_map_each *me = get_pglocked_me(&(f)->page); \
+ if (likely(me)) \
+ dept_wait_split_map(me, &pglocked_mc, _RET_IP_, \
+ __func__, 0); \
+} while (0)
+
+#define dept_pglocked_set_bit(f) \
+do { \
+ struct dept_map_each *me = get_pglocked_me(&(f)->page); \
+ if (likely(me)) \
+ dept_asked_event_split_map(me, &pglocked_mc); \
+} while (0)
+
+#define dept_pglocked_event(f) \
+do { \
+ struct dept_map_each *me = get_pglocked_me(&(f)->page); \
+ if (likely(me)) \
+ dept_event_split_map(me, &pglocked_mc, _RET_IP_,\
+ __func__); \
+} while (0)
+
+#define dept_pgwriteback_wait(f) \
+do { \
+ struct dept_map_each *me = get_pgwriteback_me(&(f)->page);\
+ if (likely(me)) \
+ dept_wait_split_map(me, &pgwriteback_mc, _RET_IP_,\
+ __func__, 0); \
+} while (0)
+
+#define dept_pgwriteback_set_bit(f) \
+do { \
+ struct dept_map_each *me = get_pgwriteback_me(&(f)->page);\
+ if (likely(me)) \
+ dept_asked_event_split_map(me, &pgwriteback_mc);\
+} while (0)
+
+#define dept_pgwriteback_event(f) \
+do { \
+ struct dept_map_each *me = get_pgwriteback_me(&(f)->page);\
+ if (likely(me)) \
+ dept_event_split_map(me, &pgwriteback_mc, _RET_IP_,\
+ __func__); \
+} while (0)
+#else
+#define dept_page_init() do { } while (0)
+#define dept_pglocked_wait(f) do { } while (0)
+#define dept_pglocked_set_bit(f) do { } while (0)
+#define dept_pglocked_event(f) do { } while (0)
+#define dept_pgwriteback_wait(f) do { } while (0)
+#define dept_pgwriteback_set_bit(f) do { } while (0)
+#define dept_pgwriteback_event(f) do { } while (0)
+#endif
+
+#endif /* __LINUX_DEPT_PAGE_H */
diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 1c3b6e5..d1c78a2 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -411,7 +411,6 @@ static unsigned long *folio_flags(struct folio *folio, unsigned n)
#define TESTSCFLAG_FALSE(uname, lname) \
TESTSETFLAG_FALSE(uname, lname) TESTCLEARFLAG_FALSE(uname, lname)
-__PAGEFLAG(Locked, locked, PF_NO_TAIL)
PAGEFLAG(Waiters, waiters, PF_ONLY_HEAD) __CLEARPAGEFLAG(Waiters, waiters, PF_ONLY_HEAD)
PAGEFLAG(Error, error, PF_NO_TAIL) TESTCLEARFLAG(Error, error, PF_NO_TAIL)
PAGEFLAG(Referenced, referenced, PF_HEAD)
@@ -459,7 +458,6 @@ static unsigned long *folio_flags(struct folio *folio, unsigned n)
* risky: they bypass page accounting.
*/
TESTPAGEFLAG(Writeback, writeback, PF_NO_TAIL)
- TESTSCFLAG(Writeback, writeback, PF_NO_TAIL)
PAGEFLAG(MappedToDisk, mappedtodisk, PF_NO_TAIL)
/* PG_readahead is only used for reads; PG_reclaim is only for writes */
@@ -542,6 +540,47 @@ static __always_inline bool PageSwapCache(struct page *page)
PAGEFLAG_FALSE(SkipKASanPoison, skip_kasan_poison)
#endif
+#ifdef CONFIG_DEPT
+TESTPAGEFLAG(Locked, locked, PF_NO_TAIL)
+__CLEARPAGEFLAG(Locked, locked, PF_NO_TAIL)
+TESTCLEARFLAG(Writeback, writeback, PF_NO_TAIL)
+
+#include <linux/dept_page.h>
+
+static __always_inline
+void __folio_set_locked(struct folio *folio)
+{
+ dept_pglocked_set_bit(folio);
+ __set_bit(PG_locked, folio_flags(folio, FOLIO_PF_NO_TAIL));
+}
+
+static __always_inline void __SetPageLocked(struct page *page)
+{
+ dept_pglocked_set_bit(page_folio(page));
+ __set_bit(PG_locked, &PF_NO_TAIL(page, 1)->flags);
+}
+
+static __always_inline
+bool folio_test_set_writeback(struct folio *folio)
+{
+ bool ret = test_and_set_bit(PG_writeback, folio_flags(folio, FOLIO_PF_NO_TAIL));
+ if (!ret)
+ dept_pgwriteback_set_bit(folio);
+ return ret;
+}
+
+static __always_inline int TestSetPageWriteback(struct page *page)
+{
+ int ret = test_and_set_bit(PG_writeback, &PF_NO_TAIL(page, 1)->flags);
+ if (!ret)
+ dept_pgwriteback_set_bit(page_folio(page));
+ return ret;
+}
+#else
+__PAGEFLAG(Locked, locked, PF_NO_TAIL)
+TESTSCFLAG(Writeback, writeback, PF_NO_TAIL)
+#endif
+
/*
* PageReported() is used to track reported free pages within the Buddy
* allocator. We can use the non-atomic version of the test and set
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 270bf51..9cfd90c 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -15,6 +15,7 @@
#include <linux/bitops.h>
#include <linux/hardirq.h> /* for in_interrupt() */
#include <linux/hugetlb_inline.h>
+#include <linux/dept_page.h>
struct folio_batch;
@@ -761,7 +762,10 @@ bool __folio_lock_or_retry(struct folio *folio, struct mm_struct *mm,
static inline bool folio_trylock(struct folio *folio)
{
- return likely(!test_and_set_bit_lock(PG_locked, folio_flags(folio, 0)));
+ int ret = test_and_set_bit_lock(PG_locked, folio_flags(folio, 0));
+ if (likely(!ret))
+ dept_pglocked_set_bit(folio);
+ return likely(!ret);
}
/*
diff --git a/init/main.c b/init/main.c
index ca96e11..4818c75 100644
--- a/init/main.c
+++ b/init/main.c
@@ -100,6 +100,7 @@
#include <linux/kcsan.h>
#include <linux/init_syscalls.h>
#include <linux/stackdepot.h>
+#include <linux/pagemap.h>
#include <asm/io.h>
#include <asm/bugs.h>
@@ -1072,6 +1073,7 @@ asmlinkage __visible void __init __no_sanitize_address start_kernel(void)
lockdep_init();
dept_init();
+ dept_page_init();
/*
* Need to run this when irqs are enabled, because it wants
diff --git a/kernel/dependency/dept_object.h b/kernel/dependency/dept_object.h
index 3e8ba81..2c4f5f0 100644
--- a/kernel/dependency/dept_object.h
+++ b/kernel/dependency/dept_object.h
@@ -5,7 +5,7 @@
* nr: # of the object that should be kept in the pool.
*/
-OBJECT(dep ,1024 * 8)
+OBJECT(dep ,1024 * 16)
OBJECT(class ,1024 * 4)
OBJECT(stack ,1024 * 32)
OBJECT(ecxt ,1024 * 4)
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 309b275..c7c2510 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1242,6 +1242,7 @@ config DEPT
select DEBUG_RWSEMS
select DEBUG_WW_MUTEX_SLOWPATH
select DEBUG_LOCK_ALLOC
+ select PAGE_EXTENSION
select TRACE_IRQFLAGS
select STACKTRACE
select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86
diff --git a/mm/filemap.c b/mm/filemap.c
index ad8c39d..16b96a4 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1148,6 +1148,11 @@ static void folio_wake_bit(struct folio *folio, int bit_nr)
unsigned long flags;
wait_queue_entry_t bookmark;
+ if (bit_nr == PG_locked)
+ dept_pglocked_event(folio);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_event(folio);
+
key.folio = folio;
key.bit_nr = bit_nr;
key.page_match = 0;
@@ -1227,6 +1232,10 @@ static inline bool folio_trylock_flag(struct folio *folio, int bit_nr,
if (wait->flags & WQ_FLAG_EXCLUSIVE) {
if (test_and_set_bit(bit_nr, &folio->flags))
return false;
+ else if (bit_nr == PG_locked)
+ dept_pglocked_set_bit(folio);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_set_bit(folio);
} else if (test_bit(bit_nr, &folio->flags))
return false;
@@ -1248,6 +1257,11 @@ static inline int folio_wait_bit_common(struct folio *folio, int bit_nr,
bool delayacct = false;
unsigned long pflags;
+ if (bit_nr == PG_locked)
+ dept_pglocked_wait(folio);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_wait(folio);
+
if (bit_nr == PG_locked &&
!folio_test_uptodate(folio) && folio_test_workingset(folio)) {
if (!folio_test_swapbacked(folio)) {
@@ -1340,6 +1354,11 @@ static inline int folio_wait_bit_common(struct folio *folio, int bit_nr,
if (unlikely(test_and_set_bit(bit_nr, folio_flags(folio, 0))))
goto repeat;
+ if (bit_nr == PG_locked)
+ dept_pglocked_set_bit(folio);
+ else if (bit_nr == PG_writeback)
+ dept_pgwriteback_set_bit(folio);
+
wait->flags |= WQ_FLAG_DONE;
break;
}
@@ -3960,3 +3979,46 @@ bool filemap_release_folio(struct folio *folio, gfp_t gfp)
return try_to_free_buffers(&folio->page);
}
EXPORT_SYMBOL(filemap_release_folio);
+
+#ifdef CONFIG_DEPT
+static bool need_dept_pglocked(void)
+{
+ return true;
+}
+
+struct page_ext_operations dept_pglocked_ops = {
+ .size = sizeof(struct dept_map_each),
+ .need = need_dept_pglocked,
+};
+
+struct dept_map_each *get_pglocked_me(struct page *p)
+{
+ struct page_ext *e = lookup_page_ext(p);
+ return e ? (void *)e + dept_pglocked_ops.offset : NULL;
+}
+
+static bool need_dept_pgwriteback(void)
+{
+ return true;
+}
+
+struct page_ext_operations dept_pgwriteback_ops = {
+ .size = sizeof(struct dept_map_each),
+ .need = need_dept_pgwriteback,
+};
+
+struct dept_map_each *get_pgwriteback_me(struct page *p)
+{
+ struct page_ext *e = lookup_page_ext(p);
+ return e ? (void *)e + dept_pgwriteback_ops.offset : NULL;
+}
+
+struct dept_map_common pglocked_mc;
+struct dept_map_common pgwriteback_mc;
+
+void dept_page_init(void)
+{
+ dept_split_map_common_init(&pglocked_mc, NULL, "pglocked");
+ dept_split_map_common_init(&pgwriteback_mc, NULL, "pgwriteback");
+}
+#endif
diff --git a/mm/page_ext.c b/mm/page_ext.c
index 2e66d93..b7f5b0d 100644
--- a/mm/page_ext.c
+++ b/mm/page_ext.c
@@ -9,6 +9,7 @@
#include <linux/page_owner.h>
#include <linux/page_idle.h>
#include <linux/page_table_check.h>
+#include <linux/dept_page.h>
/*
* struct page extension
@@ -79,6 +80,10 @@ static bool need_page_idle(void)
#ifdef CONFIG_PAGE_TABLE_CHECK
&page_table_check_ops,
#endif
+#ifdef CONFIG_DEPT
+ &dept_pglocked_ops,
+ &dept_pgwriteback_ops,
+#endif
};
unsigned long page_ext_size = sizeof(struct page_ext);
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (10 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback} Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 14/14] dept: Apply SDT to swait Byungchul Park
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Now that Dept has already been applied to major synchronization
machanisms e.g. spinlock, wait_for_completion and the like, we can take
advantage of Dept by default if CONFIG_DEPT is on.
However, sometimes we need to manually tag wait/event on places where
Dept hasn't been applied. SDT(Single-event Dependency Tracker) would be
useful in that case to tag those. The usage is like:
1. Initialize dmap in the instance associated to the wait/event.
sdt_map_init(dmap);
2. Add the following just before the wait.
sdt_wait(dmap);
3. Add the following just before the event.
sdt_event(dmap);
For better reference of SDT APIs, separated the APIs from dept.h.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/dept.h | 36 --------------------------------
include/linux/dept_sdt.h | 53 ++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 53 insertions(+), 36 deletions(-)
create mode 100644 include/linux/dept_sdt.h
diff --git a/include/linux/dept.h b/include/linux/dept.h
index 5a726bf..10ef7fc 100644
--- a/include/linux/dept.h
+++ b/include/linux/dept.h
@@ -504,32 +504,6 @@ struct dept_map_common {
*/
extern void dept_key_init(struct dept_key *k);
extern void dept_key_destroy(struct dept_key *k);
-
-#define DEPT_SDT_MAP_INIT(dname) { .name = #dname }
-#define DEFINE_DEPT_SDT(x) \
- struct dept_map x = DEPT_SDT_MAP_INIT(x)
-
-/*
- * SDT(Single-event Dependency Tracker) APIs
- *
- * In case that one dept_map instance maps to a single event, SDT APIs
- * can be used.
- */
-#define sdt_map_init(m) \
- do { \
- static struct dept_key __key; \
- dept_map_init(m, &__key, 0, #m); \
- } while (0)
-#define sdt_map_init_key(m, k) dept_map_init(m, k, 0, #m)
-
-#define sdt_wait(m) \
- do { \
- dept_asked_event(m); \
- dept_wait(m, 1UL, _THIS_IP_, "wait", 0); \
- } while (0)
-#define sdt_ecxt_enter(m) dept_ecxt_enter(m, 1UL, _THIS_IP_, "start", "event", 0)
-#define sdt_event(m) dept_event(m, 1UL, _THIS_IP_, "event")
-#define sdt_ecxt_exit(m) dept_ecxt_exit(m, _THIS_IP_)
#else /* !CONFIG_DEPT */
struct dept_task { };
struct dept_key { };
@@ -563,15 +537,5 @@ struct dept_map_common {
#define dept_asked_event_split_map(me, mc) do { } while (0)
#define dept_key_init(k) do { (void)(k); } while (0)
#define dept_key_destroy(k) do { (void)(k); } while (0)
-
-#define DEPT_SDT_MAP_INIT(dname)
-#define DEFINE_DEPT_SDT(x)
-
-#define sdt_map_init(m) do { } while (0)
-#define sdt_map_init_key(m, k) do { (void)(k); } while (0)
-#define sdt_wait(m) do { } while (0)
-#define sdt_ecxt_enter(m) do { } while (0)
-#define sdt_event(m) do { } while (0)
-#define sdt_ecxt_exit(m) do { } while (0)
#endif
#endif /* __LINUX_DEPT_H */
diff --git a/include/linux/dept_sdt.h b/include/linux/dept_sdt.h
new file mode 100644
index 0000000..32a4a9e
--- /dev/null
+++ b/include/linux/dept_sdt.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Dept Single-event Dependency Tracker
+ *
+ * Started by Byungchul Park <max.byungchul.park@gmail.com>:
+ *
+ * Copyright (c) 2020 LG Electronics, Inc., Byungchul Park
+ */
+
+#ifndef __LINUX_DEPT_SDT_H
+#define __LINUX_DEPT_SDT_H
+
+#ifdef CONFIG_DEPT
+
+#include <linux/dept.h>
+
+#define DEPT_SDT_MAP_INIT(dname) { .name = #dname }
+#define DEFINE_DEPT_SDT(x) \
+ struct dept_map x = DEPT_SDT_MAP_INIT(x)
+
+/*
+ * SDT(Single-event Dependency Tracker) APIs
+ *
+ * In case that one dept_map instance maps to a single event, SDT APIs
+ * can be used.
+ */
+#define sdt_map_init(m) \
+ do { \
+ static struct dept_key __key; \
+ dept_map_init(m, &__key, 0, #m); \
+ } while (0)
+#define sdt_map_init_key(m, k) dept_map_init(m, k, 0, #m)
+
+#define sdt_wait(m) \
+ do { \
+ dept_asked_event(m); \
+ dept_wait(m, 1UL, _THIS_IP_, "wait", 0); \
+ } while (0)
+#define sdt_ecxt_enter(m) dept_ecxt_enter(m, 1UL, _THIS_IP_, "start", "event", 0)
+#define sdt_event(m) dept_event(m, 1UL, _THIS_IP_, "event")
+#define sdt_ecxt_exit(m) dept_ecxt_exit(m, _THIS_IP_)
+#else /* !CONFIG_DEPT */
+#define DEPT_SDT_MAP_INIT(dname)
+#define DEFINE_DEPT_SDT(x)
+
+#define sdt_map_init(m) do { } while (0)
+#define sdt_map_init_key(m, k) do { (void)(k); } while (0)
+#define sdt_wait(m) do { } while (0)
+#define sdt_ecxt_enter(m) do { } while (0)
+#define sdt_event(m) do { } while (0)
+#define sdt_ecxt_exit(m) do { } while (0)
+#endif
+#endif /* __LINUX_DEPT_SDT_H */
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
* [PATCH on v5.17-rc1 14/14] dept: Apply SDT to swait
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
` (11 preceding siblings ...)
2022-01-27 1:11 ` [PATCH on v5.17-rc1 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header Byungchul Park
@ 2022-01-27 1:11 ` Byungchul Park
12 siblings, 0 replies; 31+ messages in thread
From: Byungchul Park @ 2022-01-27 1:11 UTC (permalink / raw)
To: torvalds, mingo, linux-kernel
Cc: peterz, will, tglx, rostedt, joel, sashal, daniel.vetter, chris,
duyuyang, johannes.berg, tj, tytso, willy, david, amir73il,
bfields, gregkh, kernel-team, linux-mm, akpm, mhocko, minchan,
hannes, vdavydov.dev, sj, jglisse, dennis, cl, penberg, rientjes,
vbabka, ngupta, linux-block, axboe, paolo.valente, josef,
linux-fsdevel, viro, jack, jlayton, dan.j.williams, hch, djwong,
dri-devel, airlied, rodrigosiqueiramelo, melissa.srw,
hamohammed.sa
Makes SDT able to track dependencies by swait.
Signed-off-by: Byungchul Park <byungchul.park@lge.com>
---
include/linux/swait.h | 4 ++++
kernel/sched/swait.c | 8 ++++++++
2 files changed, 12 insertions(+)
diff --git a/include/linux/swait.h b/include/linux/swait.h
index 6a8c22b..dbdf2ce 100644
--- a/include/linux/swait.h
+++ b/include/linux/swait.h
@@ -6,6 +6,7 @@
#include <linux/stddef.h>
#include <linux/spinlock.h>
#include <linux/wait.h>
+#include <linux/dept_sdt.h>
#include <asm/current.h>
/*
@@ -43,6 +44,7 @@
struct swait_queue_head {
raw_spinlock_t lock;
struct list_head task_list;
+ struct dept_map dmap;
};
struct swait_queue {
@@ -61,6 +63,7 @@ struct swait_queue {
#define __SWAIT_QUEUE_HEAD_INITIALIZER(name) { \
.lock = __RAW_SPIN_LOCK_UNLOCKED(name.lock), \
.task_list = LIST_HEAD_INIT((name).task_list), \
+ .dmap = DEPT_SDT_MAP_INIT(name), \
}
#define DECLARE_SWAIT_QUEUE_HEAD(name) \
@@ -72,6 +75,7 @@ extern void __init_swait_queue_head(struct swait_queue_head *q, const char *name
#define init_swait_queue_head(q) \
do { \
static struct lock_class_key __key; \
+ sdt_map_init(&(q)->dmap); \
__init_swait_queue_head((q), #q, &__key); \
} while (0)
diff --git a/kernel/sched/swait.c b/kernel/sched/swait.c
index e1c655f..b6c2efb 100644
--- a/kernel/sched/swait.c
+++ b/kernel/sched/swait.c
@@ -27,6 +27,7 @@ void swake_up_locked(struct swait_queue_head *q)
return;
curr = list_first_entry(&q->task_list, typeof(*curr), task_list);
+ sdt_event(&q->dmap);
wake_up_process(curr->task);
list_del_init(&curr->task_list);
}
@@ -69,6 +70,7 @@ void swake_up_all(struct swait_queue_head *q)
while (!list_empty(&tmp)) {
curr = list_first_entry(&tmp, typeof(*curr), task_list);
+ sdt_event(&q->dmap);
wake_up_state(curr->task, TASK_NORMAL);
list_del_init(&curr->task_list);
@@ -97,6 +99,9 @@ void prepare_to_swait_exclusive(struct swait_queue_head *q, struct swait_queue *
__prepare_to_swait(q, wait);
set_current_state(state);
raw_spin_unlock_irqrestore(&q->lock, flags);
+
+ if (state & TASK_NORMAL)
+ sdt_wait(&q->dmap);
}
EXPORT_SYMBOL(prepare_to_swait_exclusive);
@@ -119,6 +124,9 @@ long prepare_to_swait_event(struct swait_queue_head *q, struct swait_queue *wait
}
raw_spin_unlock_irqrestore(&q->lock, flags);
+ if (!ret && state & TASK_NORMAL)
+ sdt_wait(&q->dmap);
+
return ret;
}
EXPORT_SYMBOL(prepare_to_swait_event);
--
1.9.1
^ permalink raw reply related [flat|nested] 31+ messages in thread
end of thread, other threads:[~2022-01-27 1:12 UTC | newest]
Thread overview: 31+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-01-25 2:36 [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
2022-01-25 2:36 ` [RFC 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
2022-01-25 2:36 ` [RFC 02/14] dept: Implement Dept(Dependency Tracker) Byungchul Park
2022-01-25 2:36 ` [RFC 03/14] dept: Embed Dept data in Lockdep Byungchul Park
2022-01-25 2:36 ` [RFC 04/14] dept: Apply Dept to spinlock Byungchul Park
2022-01-25 2:36 ` [RFC 05/14] dept: Apply Dept to mutex families Byungchul Park
2022-01-25 2:36 ` [RFC 06/14] dept: Apply Dept to rwlock Byungchul Park
2022-01-25 2:36 ` [RFC 07/14] dept: Apply Dept to wait_for_completion()/complete() Byungchul Park
2022-01-25 2:36 ` [RFC 08/14] dept: Apply Dept to seqlock Byungchul Park
2022-01-25 2:36 ` [RFC 09/14] dept: Apply Dept to rwsem Byungchul Park
2022-01-25 2:36 ` [RFC 10/14] dept: Add proc knobs to show stats and dependency graph Byungchul Park
2022-01-25 2:36 ` [RFC 11/14] dept: Introduce split map concept and new APIs for them Byungchul Park
2022-01-25 2:36 ` [RFC 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback} Byungchul Park
2022-01-25 2:36 ` [RFC 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header Byungchul Park
2022-01-25 2:36 ` [RFC 14/14] dept: Apply SDT to swait Byungchul Park
2022-01-27 1:00 ` [RFC 00/14] DEPT(DEPendency Tracker) Byungchul Park
2022-01-27 1:08 ` Patches(Dept) for v5.17-rc1 Byungchul Park
2022-01-27 1:10 ` [PATCH on v5.17-rc1 01/14] llist: Move llist_{head,node} definition to types.h Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 02/14] dept: Implement Dept(Dependency Tracker) Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 03/14] dept: Embed Dept data in Lockdep Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 04/14] dept: Apply Dept to spinlock Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 05/14] dept: Apply Dept to mutex families Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 06/14] dept: Apply Dept to rwlock Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 07/14] dept: Apply Dept to wait_for_completion()/complete() Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 08/14] dept: Apply Dept to seqlock Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 09/14] dept: Apply Dept to rwsem Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 10/14] dept: Add proc knobs to show stats and dependency graph Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 11/14] dept: Introduce split map concept and new APIs for them Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 12/14] dept: Apply Dept to wait/event of PG_{locked,writeback} Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 13/14] dept: Separate out SDT(Single-event Dependency Tracker) header Byungchul Park
2022-01-27 1:11 ` [PATCH on v5.17-rc1 14/14] dept: Apply SDT to swait Byungchul Park
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).