From 778b032d96909690c19d84f8d17c13be65ed6f8e Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@redhat.com>
Date: Tue, 29 May 2012 21:30:08 +0200
Subject: uprobes: Kill uprobes_srcu/uprobe_srcu_id
Kill the no longer needed uprobes_srcu/uprobe_srcu_id code.
It doesn't really work anyway. synchronize_srcu() can only
synchronize with the code "inside" the
srcu_read_lock/srcu_read_unlock section, while
uprobe_pre_sstep_notifier() does srcu_read_lock() _after_ we
already hit the breakpoint.
I guess this probably works "in practice". synchronize_srcu() is
slow and it implies synchronize_sched(), and the probed task
enters the non- preemptible section at the start of exception
handler. Still this is not right at least in theory, and
task->uprobe_srcu_id blows task_struct.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anton Arapov <anton@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120529193008.GG8057@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
include/linux/sched.h | 1 -
1 file changed, 1 deletion(-)
(limited to 'include/linux/sched.h')
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 6029d8c5447..6bd19655c1a 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1569,7 +1569,6 @@ struct task_struct {
#endif
#ifdef CONFIG_UPROBES
struct uprobe_task *utask;
- int uprobe_srcu_id;
#endif
};
--
cgit v1.2.3-70-g09d2
From 158e1645e07f3e9f7e4962d7a0997f5c3b98311b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Jun 2012 09:24:13 +0400
Subject: trim task_work: get rid of hlist
layout based on Oleg's suggestion; single-linked list,
task->task_works points to the last element, forward pointer
from said last element points to head. I'd still prefer
much more regular scheme with two pointers in task_work,
but...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
include/linux/sched.h | 2 +-
include/linux/task_work.h | 4 +--
include/linux/tracehook.h | 2 +-
kernel/fork.c | 2 +-
kernel/task_work.c | 64 ++++++++++++++++++++++++-----------------------
5 files changed, 38 insertions(+), 36 deletions(-)
(limited to 'include/linux/sched.h')
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4059c0f33f0..b9216ebc278 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1405,7 +1405,7 @@ struct task_struct {
int (*notifier)(void *priv);
void *notifier_data;
sigset_t *notifier_mask;
- struct hlist_head task_works;
+ void *task_works;
struct audit_context *audit_context;
#ifdef CONFIG_AUDITSYSCALL
diff --git a/include/linux/task_work.h b/include/linux/task_work.h
index 627421c0e10..3b3e2c8d037 100644
--- a/include/linux/task_work.h
+++ b/include/linux/task_work.h
@@ -8,7 +8,7 @@ struct task_work;
typedef void (*task_work_func_t)(struct task_work *);
struct task_work {
- struct hlist_node hlist;
+ struct task_work *next;
task_work_func_t func;
};
@@ -24,7 +24,7 @@ void task_work_run(void);
static inline void exit_task_work(struct task_struct *task)
{
- if (unlikely(!hlist_empty(&task->task_works)))
+ if (unlikely(task->task_works))
task_work_run();
}
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 6a4d82bedb0..1e98b553042 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -192,7 +192,7 @@ static inline void tracehook_notify_resume(struct pt_regs *regs)
* hlist_add_head(task->task_works);
*/
smp_mb__after_clear_bit();
- if (unlikely(!hlist_empty(¤t->task_works)))
+ if (unlikely(current->task_works))
task_work_run();
}
diff --git a/kernel/fork.c b/kernel/fork.c
index ab5211b9e62..bebabad5920 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1415,7 +1415,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
*/
p->group_leader = p;
INIT_LIST_HEAD(&p->thread_group);
- INIT_HLIST_HEAD(&p->task_works);
+ p->task_works = NULL;
/* Now that the task is set up, run cgroup callbacks if
* necessary. We need to run them before the task is visible
diff --git a/kernel/task_work.c b/kernel/task_work.c
index 82d1c794066..9b8948dbdc6 100644
--- a/kernel/task_work.c
+++ b/kernel/task_work.c
@@ -19,7 +19,12 @@ task_work_add(struct task_struct *task, struct task_work *twork, bool notify)
*/
raw_spin_lock_irqsave(&task->pi_lock, flags);
if (likely(!(task->flags & PF_EXITING))) {
- hlist_add_head(&twork->hlist, &task->task_works);
+ struct task_work *last = task->task_works;
+ struct task_work *first = last ? last->next : twork;
+ twork->next = first;
+ if (last)
+ last->next = twork;
+ task->task_works = twork;
err = 0;
}
raw_spin_unlock_irqrestore(&task->pi_lock, flags);
@@ -34,51 +39,48 @@ struct task_work *
task_work_cancel(struct task_struct *task, task_work_func_t func)
{
unsigned long flags;
- struct task_work *twork;
- struct hlist_node *pos;
+ struct task_work *last, *res = NULL;
raw_spin_lock_irqsave(&task->pi_lock, flags);
- hlist_for_each_entry(twork, pos, &task->task_works, hlist) {
- if (twork->func == func) {
- hlist_del(&twork->hlist);
- goto found;
+ last = task->task_works;
+ if (last) {
+ struct task_work *q = last, *p = q->next;
+ while (1) {
+ if (p->func == func) {
+ q->next = p->next;
+ if (p == last)
+ task->task_works = q == p ? NULL : q;
+ res = p;
+ break;
+ }
+ if (p == last)
+ break;
+ q = p;
+ p = q->next;
}
}
- twork = NULL;
- found:
raw_spin_unlock_irqrestore(&task->pi_lock, flags);
-
- return twork;
+ return res;
}
void task_work_run(void)
{
struct task_struct *task = current;
- struct hlist_head task_works;
- struct hlist_node *pos;
+ struct task_work *p, *q;
raw_spin_lock_irq(&task->pi_lock);
- hlist_move_list(&task->task_works, &task_works);
+ p = task->task_works;
+ task->task_works = NULL;
raw_spin_unlock_irq(&task->pi_lock);
- if (unlikely(hlist_empty(&task_works)))
+ if (unlikely(!p))
return;
- /*
- * We use hlist to save the space in task_struct, but we want fifo.
- * Find the last entry, the list should be short, then process them
- * in reverse order.
- */
- for (pos = task_works.first; pos->next; pos = pos->next)
- ;
-
- for (;;) {
- struct hlist_node **pprev = pos->pprev;
- struct task_work *twork = container_of(pos, struct task_work,
- hlist);
- twork->func(twork);
- if (pprev == &task_works.first)
- break;
- pos = container_of(pprev, struct hlist_node, next);
+ q = p->next; /* head */
+ p->next = NULL; /* cut it */
+ while (q) {
+ p = q->next;
+ q->func(q);
+ q = p;
}
}
--
cgit v1.2.3-70-g09d2
From 67d1214551e800f9fe7dc7c47a346d2df0fafed5 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 27 Jun 2012 11:07:19 +0400
Subject: merge task_work and rcu_head, get rid of separate allocation for
keyring case
task_work and rcu_head are identical now; merge them (calling the result
struct callback_head, rcu_head #define'd to it), kill separate allocation
in security/keys since we can just use cred->rcu now.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
include/linux/sched.h | 2 +-
include/linux/task_work.h | 14 ++++----------
include/linux/types.h | 9 +++++----
kernel/irq/manage.c | 4 ++--
kernel/task_work.c | 14 +++++++-------
security/keys/internal.h | 6 +-----
security/keys/keyctl.c | 28 +++++++++-------------------
security/keys/process_keys.c | 6 ++----
8 files changed, 31 insertions(+), 52 deletions(-)
(limited to 'include/linux/sched.h')
diff --git a/include/linux/sched.h b/include/linux/sched.h
index b9216ebc278..af3555cc760 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1405,7 +1405,7 @@ struct task_struct {
int (*notifier)(void *priv);
void *notifier_data;
sigset_t *notifier_mask;
- void *task_works;
+ struct callback_head *task_works;
struct audit_context *audit_context;
#ifdef CONFIG_AUDITSYSCALL
diff --git a/include/linux/task_work.h b/include/linux/task_work.h
index 3b3e2c8d037..fb46b03b185 100644
--- a/include/linux/task_work.h
+++ b/include/linux/task_work.h
@@ -4,22 +4,16 @@
#include <linux/list.h>
#include <linux/sched.h>
-struct task_work;
-typedef void (*task_work_func_t)(struct task_work *);
-
-struct task_work {
- struct task_work *next;
- task_work_func_t func;
-};
+typedef void (*task_work_func_t)(struct callback_head *);
static inline void
-init_task_work(struct task_work *twork, task_work_func_t func)
+init_task_work(struct callback_head *twork, task_work_func_t func)
{
twork->func = func;
}
-int task_work_add(struct task_struct *task, struct task_work *twork, bool);
-struct task_work *task_work_cancel(struct task_struct *, task_work_func_t);
+int task_work_add(struct task_struct *task, struct callback_head *twork, bool);
+struct callback_head *task_work_cancel(struct task_struct *, task_work_func_t);
void task_work_run(void);
static inline void exit_task_work(struct task_struct *task)
diff --git a/include/linux/types.h b/include/linux/types.h
index 9c1bd539ea7..bf0dd7524b2 100644
--- a/include/linux/types.h
+++ b/include/linux/types.h
@@ -246,14 +246,15 @@ struct ustat {
};
/**
- * struct rcu_head - callback structure for use with RCU
+ * struct callback_head - callback structure for use with RCU and task_work
* @next: next update requests in a list
* @func: actual update function to call after the grace period.
*/
-struct rcu_head {
- struct rcu_head *next;
- void (*func)(struct rcu_head *head);
+struct callback_head {
+ struct callback_head *next;
+ void (*func)(struct callback_head *head);
};
+#define rcu_head callback_head
#endif /* __KERNEL__ */
#endif /* __ASSEMBLY__ */
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index d1dd54734ce..814c9ef6bba 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -781,7 +781,7 @@ static void wake_threads_waitq(struct irq_desc *desc)
wake_up(&desc->wait_for_threads);
}
-static void irq_thread_dtor(struct task_work *unused)
+static void irq_thread_dtor(struct callback_head *unused)
{
struct task_struct *tsk = current;
struct irq_desc *desc;
@@ -813,7 +813,7 @@ static void irq_thread_dtor(struct task_work *unused)
*/
static int irq_thread(void *data)
{
- struct task_work on_exit_work;
+ struct callback_head on_exit_work;
static const struct sched_param param = {
.sched_priority = MAX_USER_RT_PRIO/2,
};
diff --git a/kernel/task_work.c b/kernel/task_work.c
index 9b8948dbdc6..76266fb665d 100644
--- a/kernel/task_work.c
+++ b/kernel/task_work.c
@@ -3,7 +3,7 @@
#include <linux/tracehook.h>
int
-task_work_add(struct task_struct *task, struct task_work *twork, bool notify)
+task_work_add(struct task_struct *task, struct callback_head *twork, bool notify)
{
unsigned long flags;
int err = -ESRCH;
@@ -19,8 +19,8 @@ task_work_add(struct task_struct *task, struct task_work *twork, bool notify)
*/
raw_spin_lock_irqsave(&task->pi_lock, flags);
if (likely(!(task->flags & PF_EXITING))) {
- struct task_work *last = task->task_works;
- struct task_work *first = last ? last->next : twork;
+ struct callback_head *last = task->task_works;
+ struct callback_head *first = last ? last->next : twork;
twork->next = first;
if (last)
last->next = twork;
@@ -35,16 +35,16 @@ task_work_add(struct task_struct *task, struct task_work *twork, bool notify)
return err;
}
-struct task_work *
+struct callback_head *
task_work_cancel(struct task_struct *task, task_work_func_t func)
{
unsigned long flags;
- struct task_work *last, *res = NULL;
+ struct callback_head *last, *res = NULL;
raw_spin_lock_irqsave(&task->pi_lock, flags);
last = task->task_works;
if (last) {
- struct task_work *q = last, *p = q->next;
+ struct callback_head *q = last, *p = q->next;
while (1) {
if (p->func == func) {
q->next = p->next;
@@ -66,7 +66,7 @@ task_work_cancel(struct task_struct *task, task_work_func_t func)
void task_work_run(void)
{
struct task_struct *task = current;
- struct task_work *p, *q;
+ struct callback_head *p, *q;
raw_spin_lock_irq(&task->pi_lock);
p = task->task_works;
diff --git a/security/keys/internal.h b/security/keys/internal.h
index b510a316874..c246ba5d43a 100644
--- a/security/keys/internal.h
+++ b/security/keys/internal.h
@@ -148,12 +148,8 @@ extern key_ref_t lookup_user_key(key_serial_t id, unsigned long flags,
#define KEY_LOOKUP_PARTIAL 0x02
#define KEY_LOOKUP_FOR_UNLINK 0x04
-struct kludge { /* this will die off very soon */
- struct task_work twork;
- struct cred *cred;
-};
extern long join_session_keyring(const char *name);
-extern void key_change_session_keyring(struct task_work *twork);
+extern void key_change_session_keyring(struct callback_head *twork);
extern struct work_struct key_gc_work;
extern unsigned key_gc_delay;
diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c
index 26723caaad0..0291b3f9397 100644
--- a/security/keys/keyctl.c
+++ b/security/keys/keyctl.c
@@ -1456,8 +1456,7 @@ long keyctl_session_to_parent(void)
{
struct task_struct *me, *parent;
const struct cred *mycred, *pcred;
- struct kludge *newwork;
- struct task_work *oldwork;
+ struct callback_head *newwork, *oldwork;
key_ref_t keyring_r;
struct cred *cred;
int ret;
@@ -1467,20 +1466,17 @@ long keyctl_session_to_parent(void)
return PTR_ERR(keyring_r);
ret = -ENOMEM;
- newwork = kmalloc(sizeof(struct kludge), GFP_KERNEL);
- if (!newwork)
- goto error_keyring;
/* our parent is going to need a new cred struct, a new tgcred struct
* and new security data, so we allocate them here to prevent ENOMEM in
* our parent */
cred = cred_alloc_blank();
if (!cred)
- goto error_newwork;
+ goto error_keyring;
+ newwork = &cred->rcu;
cred->tgcred->session_keyring = key_ref_to_ptr(keyring_r);
- init_task_work(&newwork->twork, key_change_session_keyring);
- newwork->cred = cred;
+ init_task_work(newwork, key_change_session_keyring);
me = current;
rcu_read_lock();
@@ -1529,24 +1525,18 @@ long keyctl_session_to_parent(void)
/* the replacement session keyring is applied just prior to userspace
* restarting */
- ret = task_work_add(parent, &newwork->twork, true);
+ ret = task_work_add(parent, newwork, true);
if (!ret)
newwork = NULL;
unlock:
write_unlock_irq(&tasklist_lock);
rcu_read_unlock();
- if (oldwork) {
- put_cred(container_of(oldwork, struct kludge, twork)->cred);
- kfree(oldwork);
- }
- if (newwork) {
- put_cred(newwork->cred);
- kfree(newwork);
- }
+ if (oldwork)
+ put_cred(container_of(oldwork, struct cred, rcu));
+ if (newwork)
+ put_cred(cred);
return ret;
-error_newwork:
- kfree(newwork);
error_keyring:
key_ref_put(keyring_r);
return ret;
diff --git a/security/keys/process_keys.c b/security/keys/process_keys.c
index c9b07c97d7f..54339cfd673 100644
--- a/security/keys/process_keys.c
+++ b/security/keys/process_keys.c
@@ -834,13 +834,11 @@ error:
* Replace a process's session keyring on behalf of one of its children when
* the target process is about to resume userspace execution.
*/
-void key_change_session_keyring(struct task_work *twork)
+void key_change_session_keyring(struct callback_head *twork)
{
const struct cred *old = current_cred();
- struct kludge *p = container_of(twork, struct kludge, twork);
- struct cred *new = p->cred;
+ struct cred *new = container_of(twork, struct cred, rcu);
- kfree(p);
if (unlikely(current->flags & PF_EXITING)) {
put_cred(new);
return;
--
cgit v1.2.3-70-g09d2
From 6120d3dbb1220792ebea88cd475e1ec8f8620a93 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sun, 24 Jun 2012 10:03:05 +0400
Subject: get rid of ->scm_work_list
recursion in __scm_destroy() will be cut by delaying final fput()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
include/linux/sched.h | 1 -
include/net/scm.h | 1 -
net/core/scm.c | 22 +++-------------------
3 files changed, 3 insertions(+), 21 deletions(-)
(limited to 'include/linux/sched.h')
diff --git a/include/linux/sched.h b/include/linux/sched.h
index af3555cc760..598ba2da786 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1546,7 +1546,6 @@ struct task_struct {
unsigned long timer_slack_ns;
unsigned long default_timer_slack_ns;
- struct list_head *scm_work_list;
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
/* Index of current stored address in ret_stack */
int curr_ret_stack;
diff --git a/include/net/scm.h b/include/net/scm.h
index d456f4c71a3..079d7887dac 100644
--- a/include/net/scm.h
+++ b/include/net/scm.h
@@ -13,7 +13,6 @@
#define SCM_MAX_FD 253
struct scm_fp_list {
- struct list_head list;
short count;
short max;
struct file *fp[SCM_MAX_FD];
diff --git a/net/core/scm.c b/net/core/scm.c
index 611c5efd4cb..8f6ccfd68ef 100644
--- a/net/core/scm.c
+++ b/net/core/scm.c
@@ -109,25 +109,9 @@ void __scm_destroy(struct scm_cookie *scm)
if (fpl) {
scm->fp = NULL;
- if (current->scm_work_list) {
- list_add_tail(&fpl->list, current->scm_work_list);
- } else {
- LIST_HEAD(work_list);
-
- current->scm_work_list = &work_list;
-
- list_add(&fpl->list, &work_list);
- while (!list_empty(&work_list)) {
- fpl = list_first_entry(&work_list, struct scm_fp_list, list);
-
- list_del(&fpl->list);
- for (i=fpl->count-1; i>=0; i--)
- fput(fpl->fp[i]);
- kfree(fpl);
- }
-
- current->scm_work_list = NULL;
- }
+ for (i=fpl->count-1; i>=0; i--)
+ fput(fpl->fp[i]);
+ kfree(fpl);
}
}
EXPORT_SYMBOL(__scm_destroy);
--
cgit v1.2.3-70-g09d2