debian-packaging
/
linux
8
0
Fork 0
Code Releases Activity

[rt] Update to 4.9.30-rt20

This commit is contained in:
Ben Hutchings 2017-06-04 03:00:42 +01:00
parent 9c84459829
commit 894e593fa3
329 changed files with 2659 additions and 570 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
debian/changelog vendored
View File

@ -419,6 +419,22 @@ linux (4.9.30-1) UNRELEASED; urgency=medium
work again (Closes: #862723)
* [arm64] serial: pl011: add console matching function (Closes: #861898)
* [rt] Add new GPG subkeys for Sebastian Andrzej Siewior
* [rt] Update to 4.9.30-rt20:
- rtmutex: Deboost before waking up the top waiter
- sched/rtmutex/deadline: Fix a PI crash for deadline tasks
- sched/deadline/rtmutex: Dont miss the dl_runtime/dl_period update
- rtmutex: Clean up
- sched/rtmutex: Refactor rt_mutex_setprio()
- sched,tracing: Update trace_sched_pi_setprio()
- rtmutex: Fix PI chain order integrity
- rtmutex: Fix more prio comparisons
- rtmutex: Plug preempt count leak in rt_mutex_futex_unlock()
- futex: Avoid freeing an active timer
- futex: Fix small (and harmless looking) inconsistencies
- futex,rt_mutex: Fix rt_mutex_cleanup_proxy_lock()
- Revert "timers: Don't wake ktimersoftd on every tick"
- futex/rtmutex: Cure RT double blocking issue
- random: avoid preempt_disable()ed section
[ Salvatore Bonaccorso ]
* tracing: Use strlcpy() instead of strcpy() in __trace_find_cmdline()

53
debian/patches/features/all/rt/0001-futex-Avoid-freeing-an-active-timer.patch vendored Normal file
View File

@ -0,0 +1,53 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 10 Apr 2017 18:03:36 +0200
Subject: [PATCH] futex: Avoid freeing an active timer
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 97181f9bd57405b879403763284537e27d46963d
Alexander reported a hrtimer debug_object splat:
ODEBUG: free active (active state 0) object type: hrtimer hint: hrtimer_wakeup (kernel/time/hrtimer.c:1423)
debug_object_free (lib/debugobjects.c:603)
destroy_hrtimer_on_stack (kernel/time/hrtimer.c:427)
futex_lock_pi (kernel/futex.c:2740)
do_futex (kernel/futex.c:3399)
SyS_futex (kernel/futex.c:3447 kernel/futex.c:3415)
do_syscall_64 (arch/x86/entry/common.c:284)
entry_SYSCALL64_slow_path (arch/x86/entry/entry_64.S:249)
Which was caused by commit:
cfafcd117da0 ("futex: Rework futex_lock_pi() to use rt_mutex_*_proxy_lock()")
... losing the hrtimer_cancel() in the shuffle. Where previously the
hrtimer_cancel() was done by rt_mutex_slowlock() we now need to do it
manually.
Reported-by: Alexander Levin <alexander.levin@verizon.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Fixes: cfafcd117da0 ("futex: Rework futex_lock_pi() to use rt_mutex_*_proxy_lock()")
Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1704101802370.2906@nanos
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/futex.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -2734,8 +2734,10 @@ static int futex_lock_pi(u32 __user *uad
out_put_key:
put_futex_key(&q.key);
out:
- if (to)
+ if (to) {
+ hrtimer_cancel(&to->timer);
destroy_hrtimer_on_stack(&to->timer);
+ }
return ret != -EINTR ? ret : -ERESTARTNOINTR;
uaddr_faulted:

2
debian/patches/features/all/rt/0001-futex-Cleanup-variable-names-for-futex_top_waiter.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:48 +0100
Subject: [PATCH] futex: Cleanup variable names for futex_top_waiter()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 499f5aca2cdd5e958b27e2655e7e7f82524f46b1

180
debian/patches/features/all/rt/0001-rtmutex-Deboost-before-waking-up-the-top-waiter.patch vendored Normal file
View File

@ -0,0 +1,180 @@
From: Xunlei Pang <xlpang@redhat.com>
Date: Thu, 23 Mar 2017 15:56:07 +0100
Subject: [PATCH] rtmutex: Deboost before waking up the top waiter
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 2a1c6029940675abb2217b590512dbf691867ec4
We should deboost before waking the high-priority task, such that we
don't run two tasks with the same "state" (priority, deadline,
sched_class, etc).
In order to make sure the boosting task doesn't start running between
unlock and deboost (due to 'spurious' wakeup), we move the deboost
under the wait_lock, that way its serialized against the wait loop in
__rt_mutex_slowlock().
Doing the deboost early can however lead to priority-inversion if
current would get preempted after the deboost but before waking our
high-prio task, hence we disable preemption before doing deboost, and
enabling it after the wake up is over.
This gets us the right semantic order, but most importantly however;
this change ensures pointer stability for the next patch, where we
have rt_mutex_setprio() cache a pointer to the top-most waiter task.
If we, as before this change, do the wakeup first and then deboost,
this pointer might point into thin air.
[peterz: Changelog + patch munging]
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.110065320@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/futex.c | 5 ---
kernel/locking/rtmutex.c | 59 +++++++++++++++++++++-------------------
kernel/locking/rtmutex_common.h | 2 -
3 files changed, 34 insertions(+), 32 deletions(-)
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1458,10 +1458,7 @@ static int wake_futex_pi(u32 __user *uad
out_unlock:
raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
- if (deboost) {
- wake_up_q(&wake_q);
- rt_mutex_adjust_prio(current);
- }
+ rt_mutex_postunlock(&wake_q, deboost);
return ret;
}
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -371,24 +371,6 @@ static void __rt_mutex_adjust_prio(struc
}
/*
- * Adjust task priority (undo boosting). Called from the exit path of
- * rt_mutex_slowunlock() and rt_mutex_slowlock().
- *
- * (Note: We do this outside of the protection of lock->wait_lock to
- * allow the lock to be taken while or before we readjust the priority
- * of task. We do not use the spin_xx_mutex() variants here as we are
- * outside of the debug path.)
- */
-void rt_mutex_adjust_prio(struct task_struct *task)
-{
- unsigned long flags;
-
- raw_spin_lock_irqsave(&task->pi_lock, flags);
- __rt_mutex_adjust_prio(task);
- raw_spin_unlock_irqrestore(&task->pi_lock, flags);
-}
-
-/*
* Deadlock detection is conditional:
*
* If CONFIG_DEBUG_RT_MUTEXES=n, deadlock detection is only conducted
@@ -1049,6 +1031,7 @@ static void mark_wakeup_next_waiter(stru
* lock->wait_lock.
*/
rt_mutex_dequeue_pi(current, waiter);
+ __rt_mutex_adjust_prio(current);
/*
* As we are waking up the top waiter, and the waiter stays
@@ -1391,6 +1374,16 @@ static bool __sched rt_mutex_slowunlock(
*/
mark_wakeup_next_waiter(wake_q, lock);
+ /*
+ * We should deboost before waking the top waiter task such that
+ * we don't run two tasks with the 'same' priority. This however
+ * can lead to prio-inversion if we would get preempted after
+ * the deboost but before waking our high-prio task, hence the
+ * preempt_disable before unlock. Pairs with preempt_enable() in
+ * rt_mutex_postunlock();
+ */
+ preempt_disable();
+
raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
/* check PI boosting */
@@ -1440,6 +1433,18 @@ rt_mutex_fasttrylock(struct rt_mutex *lo
return slowfn(lock);
}
+/*
+ * Undo pi boosting (if necessary) and wake top waiter.
+ */
+void rt_mutex_postunlock(struct wake_q_head *wake_q, bool deboost)
+{
+ wake_up_q(wake_q);
+
+ /* Pairs with preempt_disable() in rt_mutex_slowunlock() */
+ if (deboost)
+ preempt_enable();
+}
+
static inline void
rt_mutex_fastunlock(struct rt_mutex *lock,
bool (*slowfn)(struct rt_mutex *lock,
@@ -1453,11 +1458,7 @@ rt_mutex_fastunlock(struct rt_mutex *loc
deboost = slowfn(lock, &wake_q);
- wake_up_q(&wake_q);
-
- /* Undo pi boosting if necessary: */
- if (deboost)
- rt_mutex_adjust_prio(current);
+ rt_mutex_postunlock(&wake_q, deboost);
}
/**
@@ -1570,6 +1571,13 @@ bool __sched __rt_mutex_futex_unlock(str
}
mark_wakeup_next_waiter(wake_q, lock);
+ /*
+ * We've already deboosted, retain preempt_disabled when dropping
+ * the wait_lock to avoid inversion until the wakeup. Matched
+ * by rt_mutex_postunlock();
+ */
+ preempt_disable();
+
return true; /* deboost and wakeups */
}
@@ -1582,10 +1590,7 @@ void __sched rt_mutex_futex_unlock(struc
deboost = __rt_mutex_futex_unlock(lock, &wake_q);
raw_spin_unlock_irq(&lock->wait_lock);
- if (deboost) {
- wake_up_q(&wake_q);
- rt_mutex_adjust_prio(current);
- }
+ rt_mutex_postunlock(&wake_q, deboost);
}
/**
--- a/kernel/locking/rtmutex_common.h
+++ b/kernel/locking/rtmutex_common.h
@@ -122,7 +122,7 @@ extern void rt_mutex_futex_unlock(struct
extern bool __rt_mutex_futex_unlock(struct rt_mutex *lock,
struct wake_q_head *wqh);
-extern void rt_mutex_adjust_prio(struct task_struct *task);
+extern void rt_mutex_postunlock(struct wake_q_head *wake_q, bool deboost);
#ifdef CONFIG_DEBUG_RT_MUTEXES
# include "rtmutex-debug.h"

57
debian/patches/features/all/rt/0002-futex-Fix-small-and-harmless-looking-inconsistencies.patch vendored Normal file
View File

@ -0,0 +1,57 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Fri, 7 Apr 2017 09:04:07 +0200
Subject: [PATCH] futex: Fix small (and harmless looking) inconsistencies
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 94ffac5d847cfd790bb37b7cef1cad803743985e
During (post-commit) review Darren spotted a few minor things. One
(harmless AFAICT) type inconsistency and a comment that wasn't as
clear as hoped.
Reported-by: Darren Hart (VMWare) <dvhart@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Darren Hart (VMware) <dvhart@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/futex.c | 11 +++++++----
1 file changed, 7 insertions(+), 4 deletions(-)
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1023,7 +1023,8 @@ static int attach_to_pi_state(u32 __user
struct futex_pi_state **ps)
{
pid_t pid = uval & FUTEX_TID_MASK;
- int ret, uval2;
+ u32 uval2;
+ int ret;
/*
* Userspace might have messed up non-PI and PI futexes [3]
@@ -1439,6 +1440,11 @@ static int wake_futex_pi(u32 __user *uad
if (ret)
goto out_unlock;
+ /*
+ * This is a point of no return; once we modify the uval there is no
+ * going back and subsequent operations must not fail.
+ */
+
raw_spin_lock(&pi_state->owner->pi_lock);
WARN_ON(list_empty(&pi_state->list));
list_del_init(&pi_state->list);
@@ -1450,9 +1456,6 @@ static int wake_futex_pi(u32 __user *uad
pi_state->owner = new_owner;
raw_spin_unlock(&new_owner->pi_lock);
- /*
- * We've updated the uservalue, this unlock cannot fail.
- */
postunlock = __rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q);
out_unlock:

2
debian/patches/features/all/rt/0002-futex-Use-smp_store_release-in-mark_wake_futex.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:49 +0100
Subject: [PATCH] futex: Use smp_store_release() in mark_wake_futex()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 1b367ece0d7e696cab1c8501bab282cc6a538b3f

169
debian/patches/features/all/rt/0002-sched-rtmutex-deadline-Fix-a-PI-crash-for-deadline-t.patch vendored Normal file
View File

@ -0,0 +1,169 @@
From: Xunlei Pang <xlpang@redhat.com>
Date: Thu, 23 Mar 2017 15:56:08 +0100
Subject: [PATCH] sched/rtmutex/deadline: Fix a PI crash for deadline tasks
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit e96a7705e7d3fef96aec9b590c63b2f6f7d2ba22
A crash happened while I was playing with deadline PI rtmutex.
BUG: unable to handle kernel NULL pointer dereference at 0000000000000018
IP: [<ffffffff810eeb8f>] rt_mutex_get_top_task+0x1f/0x30
PGD 232a75067 PUD 230947067 PMD 0
Oops: 0000 [#1] SMP
CPU: 1 PID: 10994 Comm: a.out Not tainted
Call Trace:
[<ffffffff810b658c>] enqueue_task+0x2c/0x80
[<ffffffff810ba763>] activate_task+0x23/0x30
[<ffffffff810d0ab5>] pull_dl_task+0x1d5/0x260
[<ffffffff810d0be6>] pre_schedule_dl+0x16/0x20
[<ffffffff8164e783>] __schedule+0xd3/0x900
[<ffffffff8164efd9>] schedule+0x29/0x70
[<ffffffff8165035b>] __rt_mutex_slowlock+0x4b/0xc0
[<ffffffff81650501>] rt_mutex_slowlock+0xd1/0x190
[<ffffffff810eeb33>] rt_mutex_timed_lock+0x53/0x60
[<ffffffff810ecbfc>] futex_lock_pi.isra.18+0x28c/0x390
[<ffffffff810ed8b0>] do_futex+0x190/0x5b0
[<ffffffff810edd50>] SyS_futex+0x80/0x180
This is because rt_mutex_enqueue_pi() and rt_mutex_dequeue_pi()
are only protected by pi_lock when operating pi waiters, while
rt_mutex_get_top_task(), will access them with rq lock held but
not holding pi_lock.
In order to tackle it, we introduce new "pi_top_task" pointer
cached in task_struct, and add new rt_mutex_update_top_task()
to update its value, it can be called by rt_mutex_setprio()
which held both owner's pi_lock and rq lock. Thus "pi_top_task"
can be safely accessed by enqueue_task_dl() under rq lock.
Originally-From: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.157682758@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/init_task.h | 1 +
include/linux/sched.h | 2 ++
include/linux/sched/rt.h | 1 +
kernel/fork.c | 1 +
kernel/locking/rtmutex.c | 29 +++++++++++++++++++++--------
kernel/sched/core.c | 2 ++
6 files changed, 28 insertions(+), 8 deletions(-)
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -164,6 +164,7 @@ extern struct task_group root_task_group
#ifdef CONFIG_RT_MUTEXES
# define INIT_RT_MUTEXES(tsk) \
.pi_waiters = RB_ROOT, \
+ .pi_top_task = NULL, \
.pi_waiters_leftmost = NULL,
#else
# define INIT_RT_MUTEXES(tsk)
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1723,6 +1723,8 @@ struct task_struct {
/* PI waiters blocked on a rt_mutex held by this task */
struct rb_root pi_waiters;
struct rb_node *pi_waiters_leftmost;
+ /* Updated under owner's pi_lock and rq lock */
+ struct task_struct *pi_top_task;
/* Deadlock detection and priority inheritance handling */
struct rt_mutex_waiter *pi_blocked_on;
#endif
--- a/include/linux/sched/rt.h
+++ b/include/linux/sched/rt.h
@@ -19,6 +19,7 @@ static inline int rt_task(struct task_st
extern int rt_mutex_getprio(struct task_struct *p);
extern void rt_mutex_setprio(struct task_struct *p, int prio);
extern int rt_mutex_get_effective_prio(struct task_struct *task, int newprio);
+extern void rt_mutex_update_top_task(struct task_struct *p);
extern struct task_struct *rt_mutex_get_top_task(struct task_struct *task);
extern void rt_mutex_adjust_pi(struct task_struct *p);
static inline bool tsk_is_pi_blocked(struct task_struct *tsk)
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1417,6 +1417,7 @@ static void rt_mutex_init_task(struct ta
#ifdef CONFIG_RT_MUTEXES
p->pi_waiters = RB_ROOT;
p->pi_waiters_leftmost = NULL;
+ p->pi_top_task = NULL;
p->pi_blocked_on = NULL;
#endif
}
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -321,6 +321,19 @@ rt_mutex_dequeue_pi(struct task_struct *
}
/*
+ * Must hold both p->pi_lock and task_rq(p)->lock.
+ */
+void rt_mutex_update_top_task(struct task_struct *p)
+{
+ if (!task_has_pi_waiters(p)) {
+ p->pi_top_task = NULL;
+ return;
+ }
+
+ p->pi_top_task = task_top_pi_waiter(p)->task;
+}
+
+/*
* Calculate task priority from the waiter tree priority
*
* Return task->normal_prio when the waiter tree is empty or when
@@ -335,12 +348,12 @@ int rt_mutex_getprio(struct task_struct
task->normal_prio);
}
+/*
+ * Must hold either p->pi_lock or task_rq(p)->lock.
+ */
struct task_struct *rt_mutex_get_top_task(struct task_struct *task)
{
- if (likely(!task_has_pi_waiters(task)))
- return NULL;
-
- return task_top_pi_waiter(task)->task;
+ return task->pi_top_task;
}
/*
@@ -349,12 +362,12 @@ struct task_struct *rt_mutex_get_top_tas
*/
int rt_mutex_get_effective_prio(struct task_struct *task, int newprio)
{
- if (!task_has_pi_waiters(task))
+ struct task_struct *top_task = rt_mutex_get_top_task(task);
+
+ if (!top_task)
return newprio;
- if (task_top_pi_waiter(task)->task->prio <= newprio)
- return task_top_pi_waiter(task)->task->prio;
- return newprio;
+ return min(top_task->prio, newprio);
}
/*
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3669,6 +3669,8 @@ void rt_mutex_setprio(struct task_struct
goto out_unlock;
}
+ rt_mutex_update_top_task(p);
+
trace_sched_pi_setprio(p, prio);
oldprio = p->prio;

38
debian/patches/features/all/rt/0003-futex-Clarify-mark_wake_futex-memory-barrier-usage.patch vendored Normal file
View File

@ -0,0 +1,38 @@
From: "Darren Hart (VMware)" <dvhart@infradead.org>
Date: Fri, 14 Apr 2017 15:31:38 -0700
Subject: [PATCH] futex: Clarify mark_wake_futex memory barrier usage
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 38fcd06e9b7f6855db1f3ebac5e18b8fdb467ffd
Clarify the scenario described in mark_wake_futex requiring the
smp_store_release(). Update the comment to explicitly refer to the
plist_del now under __unqueue_futex() (previously plist_del was in the
same function as the comment).
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170414223138.GA4222@fury
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/futex.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1378,10 +1378,11 @@ static void mark_wake_futex(struct wake_
wake_q_add(wake_q, p);
__unqueue_futex(q);
/*
- * The waiting task can free the futex_q as soon as
- * q->lock_ptr = NULL is written, without taking any locks. A
- * memory barrier is required here to prevent the following
- * store to lock_ptr from getting ahead of the plist_del.
+ * The waiting task can free the futex_q as soon as q->lock_ptr = NULL
+ * is written, without taking any locks. This is possible in the event
+ * of a spurious wakeup, for example. A memory barrier is required here
+ * to prevent the following store to lock_ptr from getting ahead of the
+ * plist_del in __unqueue_futex().
*/
smp_store_release(&q->lock_ptr, NULL);
}

2
debian/patches/features/all/rt/0003-futex-Remove-rt_mutex_deadlock_account_.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:50 +0100
Subject: [PATCH] futex: Remove rt_mutex_deadlock_account_*()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit fffa954fb528963c2fb7b0c0084eb77e2be7ab52

54
debian/patches/features/all/rt/0003-sched-deadline-rtmutex-Dont-miss-the-dl_runtime-dl_p.patch vendored Normal file
View File

@ -0,0 +1,54 @@
From: Xunlei Pang <xlpang@redhat.com>
Date: Thu, 23 Mar 2017 15:56:09 +0100
Subject: [PATCH] sched/deadline/rtmutex: Dont miss the
dl_runtime/dl_period update
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 85e2d4f992868ad78dc8bb2c077b652fcfb3661a
Currently dl tasks will actually return at the very beginning
of rt_mutex_adjust_prio_chain() in !detect_deadlock cases:
if (waiter->prio == task->prio) {
if (!detect_deadlock)
goto out_unlock_pi; // out here
else
requeue = false;
}
As the deadline value of blocked deadline tasks(waiters) without
changing their sched_class(thus prio doesn't change) never changes,
this seems reasonable, but it actually misses the chance of updating
rt_mutex_waiter's "dl_runtime(period)_copy" if a waiter updates its
deadline parameters(dl_runtime, dl_period) or boosted waiter changes
to !deadline class.
Thus, force deadline task not out by adding the !dl_prio() condition.
Signed-off-by: Xunlei Pang <xlpang@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/1460633827-345-7-git-send-email-xlpang@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.206577901@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/locking/rtmutex.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -603,7 +603,7 @@ static int rt_mutex_adjust_prio_chain(st
* enabled we continue, but stop the requeueing in the chain
* walk.
*/
- if (waiter->prio == task->prio) {
+ if (waiter->prio == task->prio && !dl_task(task)) {
if (!detect_deadlock)
goto out_unlock_pi;
else

50
debian/patches/features/all/rt/0004-MAINTAINERS-Add-FUTEX-SUBSYSTEM.patch vendored Normal file
View File

@ -0,0 +1,50 @@
From: "Darren Hart (VMware)" <dvhart@infradead.org>
Date: Fri, 14 Apr 2017 15:46:08 -0700
Subject: [PATCH] MAINTAINERS: Add FUTEX SUBSYSTEM
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 59cd42c29618c45cd3c56da43402b14f611888dd
Add a MAINTAINERS block for the FUTEX SUBSYSTEM which includes the core
kernel code, include headers, testing code, and Documentation. Excludes
arch files, and higher level test code.
I added tglx and mingo as M as they have made the tip commits and peterz
and myself as R.
Signed-off-by: Darren Hart (VMware) <dvhart@infradead.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Link: http://lkml.kernel.org/r/20170414224608.GA5180@fury
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
MAINTAINERS | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5196,6 +5196,23 @@ F: fs/fuse/
F: include/uapi/linux/fuse.h
F: Documentation/filesystems/fuse.txt
+FUTEX SUBSYSTEM
+M: Thomas Gleixner <tglx@linutronix.de>
+M: Ingo Molnar <mingo@redhat.com>
+R: Peter Zijlstra <peterz@infradead.org>
+R: Darren Hart <dvhart@infradead.org>
+L: linux-kernel@vger.kernel.org
+T: git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git locking/core
+S: Maintained
+F: kernel/futex.c
+F: kernel/futex_compat.c
+F: include/asm-generic/futex.h
+F: include/linux/futex.h
+F: include/uapi/linux/futex.h
+F: tools/testing/selftests/futex/
+F: tools/perf/bench/futex*
+F: Documentation/*futex*
+
FUTURE DOMAIN TMC-16x0 SCSI DRIVER (16-bit)
M: Rik Faith <faith@cs.unc.edu>
L: linux-scsi@vger.kernel.org

2
debian/patches/features/all/rt/0004-futex-rt_mutex-Provide-futex-specific-rt_mutex-API.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:51 +0100
Subject: [PATCH] futex,rt_mutex: Provide futex specific rt_mutex API
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 5293c2efda37775346885c7e924d4ef7018ea60b

147
debian/patches/features/all/rt/0004-rtmutex-Clean-up.patch vendored Normal file
View File

@ -0,0 +1,147 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu, 23 Mar 2017 15:56:10 +0100
Subject: [PATCH] rtmutex: Clean up
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit aa2bfe55366552cb7e93e8709d66e698d79ccc47
Previous patches changed the meaning of the return value of
rt_mutex_slowunlock(); update comments and code to reflect this.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: xlpang@redhat.com
Cc: rostedt@goodmis.org
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.255058238@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/futex.c | 7 ++++---
kernel/locking/rtmutex.c | 28 +++++++++++++---------------
kernel/locking/rtmutex_common.h | 2 +-
3 files changed, 18 insertions(+), 19 deletions(-)
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1392,7 +1392,7 @@ static int wake_futex_pi(u32 __user *uad
{
u32 uninitialized_var(curval), newval;
struct task_struct *new_owner;
- bool deboost = false;
+ bool postunlock = false;
WAKE_Q(wake_q);
int ret = 0;
@@ -1453,12 +1453,13 @@ static int wake_futex_pi(u32 __user *uad
/*
* We've updated the uservalue, this unlock cannot fail.
*/
- deboost = __rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q);
+ postunlock = __rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q);
out_unlock:
raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
- rt_mutex_postunlock(&wake_q, deboost);
+ if (postunlock)
+ rt_mutex_postunlock(&wake_q);
return ret;
}
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -1328,7 +1328,8 @@ static inline int rt_mutex_slowtrylock(s
/*
* Slow path to release a rt-mutex.
- * Return whether the current task needs to undo a potential priority boosting.
+ *
+ * Return whether the current task needs to call rt_mutex_postunlock().
*/
static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock,
struct wake_q_head *wake_q)
@@ -1399,8 +1400,7 @@ static bool __sched rt_mutex_slowunlock(
raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
- /* check PI boosting */
- return true;
+ return true; /* call rt_mutex_postunlock() */
}
/*
@@ -1447,15 +1447,14 @@ rt_mutex_fasttrylock(struct rt_mutex *lo
}
/*
- * Undo pi boosting (if necessary) and wake top waiter.
+ * Performs the wakeup of the the top-waiter and re-enables preemption.
*/
-void rt_mutex_postunlock(struct wake_q_head *wake_q, bool deboost)
+void rt_mutex_postunlock(struct wake_q_head *wake_q)
{
wake_up_q(wake_q);
/* Pairs with preempt_disable() in rt_mutex_slowunlock() */
- if (deboost)
- preempt_enable();
+ preempt_enable();
}
static inline void
@@ -1464,14 +1463,12 @@ rt_mutex_fastunlock(struct rt_mutex *loc
struct wake_q_head *wqh))
{
WAKE_Q(wake_q);
- bool deboost;
if (likely(rt_mutex_cmpxchg_release(lock, current, NULL)))
return;
- deboost = slowfn(lock, &wake_q);
-
- rt_mutex_postunlock(&wake_q, deboost);
+ if (slowfn(lock, &wake_q))
+ rt_mutex_postunlock(&wake_q);
}
/**
@@ -1591,19 +1588,20 @@ bool __sched __rt_mutex_futex_unlock(str
*/
preempt_disable();
- return true; /* deboost and wakeups */
+ return true; /* call postunlock() */
}
void __sched rt_mutex_futex_unlock(struct rt_mutex *lock)
{
WAKE_Q(wake_q);
- bool deboost;
+ bool postunlock;
raw_spin_lock_irq(&lock->wait_lock);
- deboost = __rt_mutex_futex_unlock(lock, &wake_q);
+ postunlock = __rt_mutex_futex_unlock(lock, &wake_q);
raw_spin_unlock_irq(&lock->wait_lock);
- rt_mutex_postunlock(&wake_q, deboost);
+ if (postunlock)
+ rt_mutex_postunlock(&wake_q);
}
/**
--- a/kernel/locking/rtmutex_common.h
+++ b/kernel/locking/rtmutex_common.h
@@ -122,7 +122,7 @@ extern void rt_mutex_futex_unlock(struct
extern bool __rt_mutex_futex_unlock(struct rt_mutex *lock,
struct wake_q_head *wqh);
-extern void rt_mutex_postunlock(struct wake_q_head *wake_q, bool deboost);
+extern void rt_mutex_postunlock(struct wake_q_head *wake_q);
#ifdef CONFIG_DEBUG_RT_MUTEXES
# include "rtmutex-debug.h"

2
debian/patches/features/all/rt/0005-futex-Change-locking-rules.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:52 +0100
Subject: [PATCH] futex: Change locking rules
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 734009e96d1983ad739e5b656e03430b3660c913

393
debian/patches/features/all/rt/0005-sched-rtmutex-Refactor-rt_mutex_setprio.patch vendored Normal file
View File

@ -0,0 +1,393 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu, 23 Mar 2017 15:56:11 +0100
Subject: [PATCH] sched/rtmutex: Refactor rt_mutex_setprio()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit acd58620e415aee4a43a808d7d2fd87259ee0001
With the introduction of SCHED_DEADLINE the whole notion that priority
is a single number is gone, therefore the @prio argument to
rt_mutex_setprio() doesn't make sense anymore.
So rework the code to pass a pi_task instead.
Note this also fixes a problem with pi_top_task caching; previously we
would not set the pointer (call rt_mutex_update_top_task) if the
priority didn't change, this could lead to a stale pointer.
As for the XXX, I think its fine to use pi_task->prio, because if it
differs from waiter->prio, a PI chain update is immenent.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: xlpang@redhat.com
Cc: rostedt@goodmis.org
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.303827095@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/sched/rt.h | 24 +++-------
kernel/locking/rtmutex.c | 112 ++++++++++++-----------------------------------
kernel/sched/core.c | 66 ++++++++++++++++++++++-----
3 files changed, 91 insertions(+), 111 deletions(-)
--- a/include/linux/sched/rt.h
+++ b/include/linux/sched/rt.h
@@ -16,28 +16,20 @@ static inline int rt_task(struct task_st
}
#ifdef CONFIG_RT_MUTEXES
-extern int rt_mutex_getprio(struct task_struct *p);
-extern void rt_mutex_setprio(struct task_struct *p, int prio);
-extern int rt_mutex_get_effective_prio(struct task_struct *task, int newprio);
-extern void rt_mutex_update_top_task(struct task_struct *p);
-extern struct task_struct *rt_mutex_get_top_task(struct task_struct *task);
+/*
+ * Must hold either p->pi_lock or task_rq(p)->lock.
+ */
+static inline struct task_struct *rt_mutex_get_top_task(struct task_struct *p)
+{
+ return p->pi_top_task;
+}
+extern void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task);
extern void rt_mutex_adjust_pi(struct task_struct *p);
static inline bool tsk_is_pi_blocked(struct task_struct *tsk)
{
return tsk->pi_blocked_on != NULL;
}
#else
-static inline int rt_mutex_getprio(struct task_struct *p)
-{
- return p->normal_prio;
-}
-
-static inline int rt_mutex_get_effective_prio(struct task_struct *task,
- int newprio)
-{
- return newprio;
-}
-
static inline struct task_struct *rt_mutex_get_top_task(struct task_struct *task)
{
return NULL;
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -320,67 +320,16 @@ rt_mutex_dequeue_pi(struct task_struct *
RB_CLEAR_NODE(&waiter->pi_tree_entry);
}
-/*
- * Must hold both p->pi_lock and task_rq(p)->lock.
- */
-void rt_mutex_update_top_task(struct task_struct *p)
-{
- if (!task_has_pi_waiters(p)) {
- p->pi_top_task = NULL;
- return;
- }
-
- p->pi_top_task = task_top_pi_waiter(p)->task;
-}
-
-/*
- * Calculate task priority from the waiter tree priority
- *
- * Return task->normal_prio when the waiter tree is empty or when
- * the waiter is not allowed to do priority boosting
- */
-int rt_mutex_getprio(struct task_struct *task)
-{
- if (likely(!task_has_pi_waiters(task)))
- return task->normal_prio;
-
- return min(task_top_pi_waiter(task)->prio,
- task->normal_prio);
-}
-
-/*
- * Must hold either p->pi_lock or task_rq(p)->lock.
- */
-struct task_struct *rt_mutex_get_top_task(struct task_struct *task)
-{
- return task->pi_top_task;
-}
-
-/*
- * Called by sched_setscheduler() to get the priority which will be
- * effective after the change.
- */
-int rt_mutex_get_effective_prio(struct task_struct *task, int newprio)
+static void rt_mutex_adjust_prio(struct task_struct *p)
{
- struct task_struct *top_task = rt_mutex_get_top_task(task);
+ struct task_struct *pi_task = NULL;
- if (!top_task)
- return newprio;
+ lockdep_assert_held(&p->pi_lock);
- return min(top_task->prio, newprio);
-}
+ if (task_has_pi_waiters(p))
+ pi_task = task_top_pi_waiter(p)->task;
-/*
- * Adjust the priority of a task, after its pi_waiters got modified.
- *
- * This can be both boosting and unboosting. task->pi_lock must be held.
- */
-static void __rt_mutex_adjust_prio(struct task_struct *task)
-{
- int prio = rt_mutex_getprio(task);
-
- if (task->prio != prio || dl_prio(prio))
- rt_mutex_setprio(task, prio);
+ rt_mutex_setprio(p, pi_task);
}
/*
@@ -740,7 +689,7 @@ static int rt_mutex_adjust_prio_chain(st
*/
rt_mutex_dequeue_pi(task, prerequeue_top_waiter);
rt_mutex_enqueue_pi(task, waiter);
- __rt_mutex_adjust_prio(task);
+ rt_mutex_adjust_prio(task);
} else if (prerequeue_top_waiter == waiter) {
/*
@@ -756,7 +705,7 @@ static int rt_mutex_adjust_prio_chain(st
rt_mutex_dequeue_pi(task, waiter);
waiter = rt_mutex_top_waiter(lock);
rt_mutex_enqueue_pi(task, waiter);
- __rt_mutex_adjust_prio(task);
+ rt_mutex_adjust_prio(task);
} else {
/*
* Nothing changed. No need to do any priority
@@ -964,7 +913,7 @@ static int task_blocks_on_rt_mutex(struc
return -EDEADLK;
raw_spin_lock(&task->pi_lock);
- __rt_mutex_adjust_prio(task);
+ rt_mutex_adjust_prio(task);
waiter->task = task;
waiter->lock = lock;
waiter->prio = task->prio;
@@ -986,7 +935,7 @@ static int task_blocks_on_rt_mutex(struc
rt_mutex_dequeue_pi(owner, top_waiter);
rt_mutex_enqueue_pi(owner, waiter);
- __rt_mutex_adjust_prio(owner);
+ rt_mutex_adjust_prio(owner);
if (owner->pi_blocked_on)
chain_walk = 1;
} else if (rt_mutex_cond_detect_deadlock(waiter, chwalk)) {
@@ -1038,13 +987,14 @@ static void mark_wakeup_next_waiter(stru
waiter = rt_mutex_top_waiter(lock);
/*
- * Remove it from current->pi_waiters. We do not adjust a
- * possible priority boost right now. We execute wakeup in the
- * boosted mode and go back to normal after releasing
- * lock->wait_lock.
+ * Remove it from current->pi_waiters and deboost.
+ *
+ * We must in fact deboost here in order to ensure we call
+ * rt_mutex_setprio() to update p->pi_top_task before the
+ * task unblocks.
*/
rt_mutex_dequeue_pi(current, waiter);
- __rt_mutex_adjust_prio(current);
+ rt_mutex_adjust_prio(current);
/*
* As we are waking up the top waiter, and the waiter stays
@@ -1056,9 +1006,19 @@ static void mark_wakeup_next_waiter(stru
*/
lock->owner = (void *) RT_MUTEX_HAS_WAITERS;
- raw_spin_unlock(&current->pi_lock);
-
+ /*
+ * We deboosted before waking the top waiter task such that we don't
+ * run two tasks with the 'same' priority (and ensure the
+ * p->pi_top_task pointer points to a blocked task). This however can
+ * lead to priority inversion if we would get preempted after the
+ * deboost but before waking our donor task, hence the preempt_disable()
+ * before unlock.
+ *
+ * Pairs with preempt_enable() in rt_mutex_postunlock();
+ */
+ preempt_disable();
wake_q_add(wake_q, waiter->task);
+ raw_spin_unlock(&current->pi_lock);
}
/*
@@ -1093,7 +1053,7 @@ static void remove_waiter(struct rt_mute
if (rt_mutex_has_waiters(lock))
rt_mutex_enqueue_pi(owner, rt_mutex_top_waiter(lock));
- __rt_mutex_adjust_prio(owner);
+ rt_mutex_adjust_prio(owner);
/* Store the lock on which owner is blocked or NULL */
next_lock = task_blocked_on_lock(owner);
@@ -1132,8 +1092,7 @@ void rt_mutex_adjust_pi(struct task_stru
raw_spin_lock_irqsave(&task->pi_lock, flags);
waiter = task->pi_blocked_on;
- if (!waiter || (waiter->prio == task->prio &&
- !dl_prio(task->prio))) {
+ if (!waiter || (waiter->prio == task->prio && !dl_prio(task->prio))) {
raw_spin_unlock_irqrestore(&task->pi_lock, flags);
return;
}
@@ -1387,17 +1346,6 @@ static bool __sched rt_mutex_slowunlock(
* Queue the next waiter for wakeup once we release the wait_lock.
*/
mark_wakeup_next_waiter(wake_q, lock);
-
- /*
- * We should deboost before waking the top waiter task such that
- * we don't run two tasks with the 'same' priority. This however
- * can lead to prio-inversion if we would get preempted after
- * the deboost but before waking our high-prio task, hence the
- * preempt_disable before unlock. Pairs with preempt_enable() in
- * rt_mutex_postunlock();
- */
- preempt_disable();
-
raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
return true; /* call rt_mutex_postunlock() */
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3629,10 +3629,25 @@ EXPORT_SYMBOL(default_wake_function);
#ifdef CONFIG_RT_MUTEXES
+static inline int __rt_effective_prio(struct task_struct *pi_task, int prio)
+{
+ if (pi_task)
+ prio = min(prio, pi_task->prio);
+
+ return prio;
+}
+
+static inline int rt_effective_prio(struct task_struct *p, int prio)
+{
+ struct task_struct *pi_task = rt_mutex_get_top_task(p);
+
+ return __rt_effective_prio(pi_task, prio);
+}
+
/*
* rt_mutex_setprio - set the current priority of a task
- * @p: task
- * @prio: prio value (kernel-internal form)
+ * @p: task to boost
+ * @pi_task: donor task
*
* This function changes the 'effective' priority of a task. It does
* not touch ->normal_prio like __setscheduler().
@@ -3640,16 +3655,40 @@ EXPORT_SYMBOL(default_wake_function);
* Used by the rt_mutex code to implement priority inheritance
* logic. Call site only calls if the priority of the task changed.
*/
-void rt_mutex_setprio(struct task_struct *p, int prio)
+void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task)
{
- int oldprio, queued, running, queue_flag = DEQUEUE_SAVE | DEQUEUE_MOVE;
+ int prio, oldprio, queued, running, queue_flag = DEQUEUE_SAVE | DEQUEUE_MOVE;
const struct sched_class *prev_class;
struct rq_flags rf;
struct rq *rq;
- BUG_ON(prio > MAX_PRIO);
+ /* XXX used to be waiter->prio, not waiter->task->prio */
+ prio = __rt_effective_prio(pi_task, p->normal_prio);
+
+ /*
+ * If nothing changed; bail early.
+ */
+ if (p->pi_top_task == pi_task && prio == p->prio && !dl_prio(prio))
+ return;
rq = __task_rq_lock(p, &rf);
+ /*
+ * Set under pi_lock && rq->lock, such that the value can be used under
+ * either lock.
+ *
+ * Note that there is loads of tricky to make this pointer cache work
+ * right. rt_mutex_slowunlock()+rt_mutex_postunlock() work together to
+ * ensure a task is de-boosted (pi_task is set to NULL) before the
+ * task is allowed to run again (and can exit). This ensures the pointer
+ * points to a blocked task -- which guaratees the task is present.
+ */
+ p->pi_top_task = pi_task;
+
+ /*
+ * For FIFO/RR we only need to set prio, if that matches we're done.
+ */
+ if (prio == p->prio && !dl_prio(prio))
+ goto out_unlock;
/*
* Idle task boosting is a nono in general. There is one
@@ -3669,9 +3708,7 @@ void rt_mutex_setprio(struct task_struct
goto out_unlock;
}
- rt_mutex_update_top_task(p);
-
- trace_sched_pi_setprio(p, prio);
+ trace_sched_pi_setprio(p, prio); /* broken */
oldprio = p->prio;
if (oldprio == prio)
@@ -3695,7 +3732,6 @@ void rt_mutex_setprio(struct task_struct
* running task
*/
if (dl_prio(prio)) {
- struct task_struct *pi_task = rt_mutex_get_top_task(p);
if (!dl_prio(p->normal_prio) ||
(pi_task && dl_entity_preempt(&pi_task->dl, &p->dl))) {
p->dl.dl_boosted = 1;
@@ -3732,6 +3768,11 @@ void rt_mutex_setprio(struct task_struct
balance_callback(rq);
preempt_enable();
}
+#else
+static inline int rt_effective_prio(struct task_struct *p, int prio)
+{
+ return prio;
+}
#endif
void set_user_nice(struct task_struct *p, long nice)
@@ -3976,10 +4017,9 @@ static void __setscheduler(struct rq *rq
* Keep a potential priority boosting if called from
* sched_setscheduler().
*/
+ p->prio = normal_prio(p);
if (keep_boost)
- p->prio = rt_mutex_get_effective_prio(p, normal_prio(p));
- else
- p->prio = normal_prio(p);
+ p->prio = rt_effective_prio(p, p->prio);
if (dl_prio(p->prio))
p->sched_class = &dl_sched_class;
@@ -4266,7 +4306,7 @@ static int __sched_setscheduler(struct t
* the runqueue. This will be done when the task deboost
* itself.
*/
- new_effective_prio = rt_mutex_get_effective_prio(p, newprio);
+ new_effective_prio = rt_effective_prio(p, newprio);
if (new_effective_prio == oldprio)
queue_flags &= ~DEQUEUE_MOVE;
}

2
debian/patches/features/all/rt/0006-futex-Cleanup-refcounting.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:53 +0100
Subject: [PATCH] futex: Cleanup refcounting
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit bf92cf3a5100f5a0d5f9834787b130159397cb22

109
debian/patches/features/all/rt/0006-sched-tracing-Update-trace_sched_pi_setprio.patch vendored Normal file
View File

@ -0,0 +1,109 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu, 23 Mar 2017 15:56:12 +0100
Subject: [PATCH] sched,tracing: Update trace_sched_pi_setprio()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit b91473ff6e979c0028f02f90e40c844959c736d8
Pass the PI donor task, instead of a numerical priority.
Numerical priorities are not sufficient to describe state ever since
SCHED_DEADLINE.
Annotate all sched tracepoints that are currently broken; fixing them
will bork userspace. *hate*.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: xlpang@redhat.com
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.353599881@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/trace/events/sched.h | 16 +++++++++-------
kernel/sched/core.c | 2 +-
2 files changed, 10 insertions(+), 8 deletions(-)
--- a/include/trace/events/sched.h
+++ b/include/trace/events/sched.h
@@ -70,7 +70,7 @@ DECLARE_EVENT_CLASS(sched_wakeup_templat
TP_fast_assign(
memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
__entry->pid = p->pid;
- __entry->prio = p->prio;
+ __entry->prio = p->prio; /* XXX SCHED_DEADLINE */
__entry->success = 1; /* rudiment, kill when possible */
__entry->target_cpu = task_cpu(p);
),
@@ -147,6 +147,7 @@ TRACE_EVENT(sched_switch,
memcpy(__entry->prev_comm, prev->comm, TASK_COMM_LEN);
__entry->next_pid = next->pid;
__entry->next_prio = next->prio;
+ /* XXX SCHED_DEADLINE */
),
TP_printk("prev_comm=%s prev_pid=%d prev_prio=%d prev_state=%s%s ==> next_comm=%s next_pid=%d next_prio=%d",
@@ -181,7 +182,7 @@ TRACE_EVENT(sched_migrate_task,
TP_fast_assign(
memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
__entry->pid = p->pid;
- __entry->prio = p->prio;
+ __entry->prio = p->prio; /* XXX SCHED_DEADLINE */
__entry->orig_cpu = task_cpu(p);
__entry->dest_cpu = dest_cpu;
),
@@ -206,7 +207,7 @@ DECLARE_EVENT_CLASS(sched_process_templa
TP_fast_assign(
memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
__entry->pid = p->pid;
- __entry->prio = p->prio;
+ __entry->prio = p->prio; /* XXX SCHED_DEADLINE */
),
TP_printk("comm=%s pid=%d prio=%d",
@@ -253,7 +254,7 @@ TRACE_EVENT(sched_process_wait,
TP_fast_assign(
memcpy(__entry->comm, current->comm, TASK_COMM_LEN);
__entry->pid = pid_nr(pid);
- __entry->prio = current->prio;
+ __entry->prio = current->prio; /* XXX SCHED_DEADLINE */
),
TP_printk("comm=%s pid=%d prio=%d",
@@ -413,9 +414,9 @@ DEFINE_EVENT(sched_stat_runtime, sched_s
*/
TRACE_EVENT(sched_pi_setprio,
- TP_PROTO(struct task_struct *tsk, int newprio),
+ TP_PROTO(struct task_struct *tsk, struct task_struct *pi_task),
- TP_ARGS(tsk, newprio),
+ TP_ARGS(tsk, pi_task),
TP_STRUCT__entry(
__array( char, comm, TASK_COMM_LEN )
@@ -428,7 +429,8 @@ TRACE_EVENT(sched_pi_setprio,
memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
__entry->pid = tsk->pid;
__entry->oldprio = tsk->prio;
- __entry->newprio = newprio;
+ __entry->newprio = pi_task ? pi_task->prio : tsk->prio;
+ /* XXX SCHED_DEADLINE bits missing */
),
TP_printk("comm=%s pid=%d oldprio=%d newprio=%d",
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3708,7 +3708,7 @@ void rt_mutex_setprio(struct task_struct
goto out_unlock;
}
- trace_sched_pi_setprio(p, prio); /* broken */
+ trace_sched_pi_setprio(p, pi_task);
oldprio = p->prio;
if (oldprio == prio)

2
debian/patches/features/all/rt/0007-futex-Rework-inconsistent-rt_mutex-futex_q-state.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:54 +0100
Subject: [PATCH] futex: Rework inconsistent rt_mutex/futex_q state
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 73d786bd043ebc855f349c81ea805f6b11cbf2aa

122
debian/patches/features/all/rt/0007-rtmutex-Fix-PI-chain-order-integrity.patch vendored Normal file
View File

@ -0,0 +1,122 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu, 23 Mar 2017 15:56:13 +0100
Subject: [PATCH] rtmutex: Fix PI chain order integrity
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit e0aad5b44ff5d28ac1d6ae70cdf84ca228e889dc
rt_mutex_waiter::prio is a copy of task_struct::prio which is updated
during the PI chain walk, such that the PI chain order isn't messed up
by (asynchronous) task state updates.
Currently rt_mutex_waiter_less() uses task state for deadline tasks;
this is broken, since the task state can, as said above, change
asynchronously, causing the RB tree order to change without actual
tree update -> FAIL.
Fix this by also copying the deadline into the rt_mutex_waiter state
and updating it along with its prio field.
Ideally we would also force PI chain updates whenever DL tasks update
their deadline parameter, but for first approximation this is less
broken than it was.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: xlpang@redhat.com
Cc: rostedt@goodmis.org
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.403992539@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/locking/rtmutex.c | 29 +++++++++++++++++++++++++++--
kernel/locking/rtmutex_common.h | 1 +
2 files changed, 28 insertions(+), 2 deletions(-)
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -236,8 +236,7 @@ rt_mutex_waiter_less(struct rt_mutex_wai
* then right waiter has a dl_prio() too.
*/
if (dl_prio(left->prio))
- return dl_time_before(left->task->dl.deadline,
- right->task->dl.deadline);
+ return dl_time_before(left->deadline, right->deadline);
return 0;
}
@@ -648,7 +647,26 @@ static int rt_mutex_adjust_prio_chain(st
/* [7] Requeue the waiter in the lock waiter tree. */
rt_mutex_dequeue(lock, waiter);
+
+ /*
+ * Update the waiter prio fields now that we're dequeued.
+ *
+ * These values can have changed through either:
+ *
+ * sys_sched_set_scheduler() / sys_sched_setattr()
+ *
+ * or
+ *
+ * DL CBS enforcement advancing the effective deadline.
+ *
+ * Even though pi_waiters also uses these fields, and that tree is only
+ * updated in [11], we can do this here, since we hold [L], which
+ * serializes all pi_waiters access and rb_erase() does not care about
+ * the values of the node being removed.
+ */
waiter->prio = task->prio;
+ waiter->deadline = task->dl.deadline;
+
rt_mutex_enqueue(lock, waiter);
/* [8] Release the task */
@@ -775,6 +793,8 @@ static int rt_mutex_adjust_prio_chain(st
static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task,
struct rt_mutex_waiter *waiter)
{
+ lockdep_assert_held(&lock->wait_lock);
+
/*
* Before testing whether we can acquire @lock, we set the
* RT_MUTEX_HAS_WAITERS bit in @lock->owner. This forces all
@@ -900,6 +920,8 @@ static int task_blocks_on_rt_mutex(struc
struct rt_mutex *next_lock;
int chain_walk = 0, res;
+ lockdep_assert_held(&lock->wait_lock);
+
/*
* Early deadlock detection. We really don't want the task to
* enqueue on itself just to untangle the mess later. It's not
@@ -917,6 +939,7 @@ static int task_blocks_on_rt_mutex(struc
waiter->task = task;
waiter->lock = lock;
waiter->prio = task->prio;
+ waiter->deadline = task->dl.deadline;
/* Get the top priority waiter on the lock */
if (rt_mutex_has_waiters(lock))
@@ -1034,6 +1057,8 @@ static void remove_waiter(struct rt_mute
struct task_struct *owner = rt_mutex_owner(lock);
struct rt_mutex *next_lock;
+ lockdep_assert_held(&lock->wait_lock);
+
raw_spin_lock(&current->pi_lock);
rt_mutex_dequeue(lock, waiter);
current->pi_blocked_on = NULL;
--- a/kernel/locking/rtmutex_common.h
+++ b/kernel/locking/rtmutex_common.h
@@ -33,6 +33,7 @@ struct rt_mutex_waiter {
struct rt_mutex *deadlock_lock;
#endif
int prio;
+ u64 deadline;
};
/*

2
debian/patches/features/all/rt/0008-futex-Pull-rt_mutex_futex_unlock-out-from-under-hb-l.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:55 +0100
Subject: [PATCH] futex: Pull rt_mutex_futex_unlock() out from under hb->lock
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 16ffa12d742534d4ff73e8b3a4e81c1de39196f0

102
debian/patches/features/all/rt/0008-rtmutex-Fix-more-prio-comparisons.patch vendored Normal file
View File

@ -0,0 +1,102 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu, 23 Mar 2017 15:56:14 +0100
Subject: [PATCH] rtmutex: Fix more prio comparisons
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 19830e55247cddb3f46f1bf60b8e245593491bea
There was a pure ->prio comparison left in try_to_wake_rt_mutex(),
convert it to use rt_mutex_waiter_less(), noting that greater-or-equal
is not-less (both in kernel priority view).
This necessitated the introduction of cmp_task() which creates a
pointer to an unnamed stack variable of struct rt_mutex_waiter type to
compare against tasks.
With this, we can now also create and employ rt_mutex_waiter_equal().
Reviewed-and-tested-by: Juri Lelli <juri.lelli@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: juri.lelli@arm.com
Cc: bigeasy@linutronix.de
Cc: xlpang@redhat.com
Cc: rostedt@goodmis.org
Cc: mathieu.desnoyers@efficios.com
Cc: jdesfossez@efficios.com
Cc: bristot@redhat.com
Link: http://lkml.kernel.org/r/20170323150216.455584638@infradead.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/locking/rtmutex.c | 32 +++++++++++++++++++++++++++++---
1 file changed, 29 insertions(+), 3 deletions(-)
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -222,6 +222,12 @@ static inline bool unlock_rt_mutex_safe(
}
#endif
+/*
+ * Only use with rt_mutex_waiter_{less,equal}()
+ */
+#define task_to_waiter(p) \
+ &(struct rt_mutex_waiter){ .prio = (p)->prio, .deadline = (p)->dl.deadline }
+
static inline int
rt_mutex_waiter_less(struct rt_mutex_waiter *left,
struct rt_mutex_waiter *right)
@@ -241,6 +247,25 @@ rt_mutex_waiter_less(struct rt_mutex_wai
return 0;
}
+static inline int
+rt_mutex_waiter_equal(struct rt_mutex_waiter *left,
+ struct rt_mutex_waiter *right)
+{
+ if (left->prio != right->prio)
+ return 0;
+
+ /*
+ * If both waiters have dl_prio(), we check the deadlines of the
+ * associated tasks.
+ * If left waiter has a dl_prio(), and we didn't return 0 above,
+ * then right waiter has a dl_prio() too.
+ */
+ if (dl_prio(left->prio))
+ return left->deadline == right->deadline;
+
+ return 1;
+}
+
static void
rt_mutex_enqueue(struct rt_mutex *lock, struct rt_mutex_waiter *waiter)
{
@@ -551,7 +576,7 @@ static int rt_mutex_adjust_prio_chain(st
* enabled we continue, but stop the requeueing in the chain
* walk.
*/
- if (waiter->prio == task->prio && !dl_task(task)) {
+ if (rt_mutex_waiter_equal(waiter, task_to_waiter(task))) {
if (!detect_deadlock)
goto out_unlock_pi;
else
@@ -854,7 +879,8 @@ static int try_to_take_rt_mutex(struct r
* the top waiter priority (kernel view),
* @task lost.
*/
- if (task->prio >= rt_mutex_top_waiter(lock)->prio)
+ if (!rt_mutex_waiter_less(task_to_waiter(task),
+ rt_mutex_top_waiter(lock)))
return 0;
/*
@@ -1117,7 +1143,7 @@ void rt_mutex_adjust_pi(struct task_stru
raw_spin_lock_irqsave(&task->pi_lock, flags);
waiter = task->pi_blocked_on;
- if (!waiter || (waiter->prio == task->prio && !dl_prio(task->prio))) {
+ if (!waiter || rt_mutex_waiter_equal(waiter, task_to_waiter(task))) {
raw_spin_unlock_irqrestore(&task->pi_lock, flags);
return;
}

2
debian/patches/features/all/rt/0009-futex-rt_mutex-Introduce-rt_mutex_init_waiter.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:56 +0100
Subject: [PATCH] futex,rt_mutex: Introduce rt_mutex_init_waiter()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 50809358dd7199aa7ce232f6877dd09ec30ef374

43
debian/patches/features/all/rt/0009-rtmutex-Plug-preempt-count-leak-in-rt_mutex_futex_un.patch vendored Normal file
View File

@ -0,0 +1,43 @@
From: Mike Galbraith <efault@gmx.de>
Date: Wed, 5 Apr 2017 10:08:27 +0200
Subject: [PATCH] rtmutex: Plug preempt count leak in
rt_mutex_futex_unlock()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit def34eaae5ce04b324e48e1bfac873091d945213
mark_wakeup_next_waiter() already disables preemption, doing so again
leaves us with an unpaired preempt_disable().
Fixes: 2a1c60299406 ("rtmutex: Deboost before waking up the top waiter")
Signed-off-by: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: xlpang@redhat.com
Cc: rostedt@goodmis.org
Link: http://lkml.kernel.org/r/1491379707.6538.2.camel@gmx.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/locking/rtmutex.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
--- a/kernel/locking/rtmutex.c
+++ b/kernel/locking/rtmutex.c
@@ -1579,13 +1579,13 @@ bool __sched __rt_mutex_futex_unlock(str
return false; /* done */
}
- mark_wakeup_next_waiter(wake_q, lock);
/*
- * We've already deboosted, retain preempt_disabled when dropping
- * the wait_lock to avoid inversion until the wakeup. Matched
- * by rt_mutex_postunlock();
+ * We've already deboosted, mark_wakeup_next_waiter() will
+ * retain preempt_disabled when we drop the wait_lock, to
+ * avoid inversion prior to the wakeup. preempt_disable()
+ * therein pairs with rt_mutex_postunlock().
*/
- preempt_disable();
+ mark_wakeup_next_waiter(wake_q, lock);
return true; /* call postunlock() */
}

2
debian/patches/features/all/rt/0010-futex-rt_mutex-Restructure-rt_mutex_finish_proxy_loc.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:57 +0100
Subject: [PATCH] futex,rt_mutex: Restructure rt_mutex_finish_proxy_lock()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 38d589f2fd08f1296aea3ce62bebd185125c6d81

2
debian/patches/features/all/rt/0011-futex-Rework-futex_lock_pi-to-use-rt_mutex_-_proxy_l.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:58 +0100
Subject: [PATCH] futex: Rework futex_lock_pi() to use rt_mutex_*_proxy_lock()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit cfafcd117da0216520568c195cb2f6cd1980c4bb

2
debian/patches/features/all/rt/0012-futex-Futex_unlock_pi-determinism.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:35:59 +0100
Subject: [PATCH] futex: Futex_unlock_pi() determinism
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit bebe5b514345f09be2c15e414d076b02ecb9cce8

2
debian/patches/features/all/rt/0013-futex-Drop-hb-lock-before-enqueueing-on-the-rtmutex.patch vendored
View File

@ -1,7 +1,7 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed, 22 Mar 2017 11:36:00 +0100
Subject: [PATCH] futex: Drop hb->lock before enqueueing on the rtmutex
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Upstream commit 56222b212e8edb1cf51f5dd73ff645809b082b40

2
debian/patches/features/all/rt/ARM-enable-irq-in-translation-section-permission-fau.patch vendored
View File

@ -1,7 +1,7 @@
From: "Yadi.hu" <yadi.hu@windriver.com>
Date: Wed, 10 Dec 2014 10:32:09 +0800
Subject: ARM: enable irq in translation/section permission fault handlers
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Probably happens on all ARM, with
CONFIG_PREEMPT_RT_FULL

2
debian/patches/features/all/rt/HACK-printk-drop-the-logbuf_lock-more-often.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 21 Mar 2013 19:01:05 +0100
Subject: printk: Drop the logbuf_lock more often
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The lock is hold with irgs off. The latency drops 500us+ on my arm bugs
with a "full" buffer after executing "dmesg" on the shell.

2
debian/patches/features/all/rt/KVM-arm-arm64-downgrade-preempt_disable-d-region-to-.patch vendored
View File

@ -1,7 +1,7 @@
From: Josh Cartwright <joshc@ni.com>
Date: Thu, 11 Feb 2016 11:54:01 -0600
Subject: KVM: arm/arm64: downgrade preempt_disable()d region to migrate_disable()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
kvm_arch_vcpu_ioctl_run() disables the use of preemption when updating
the vgic and timer states to prevent the calling task from migrating to

2
debian/patches/features/all/rt/KVM-lapic-mark-LAPIC-timer-handler-as-irqsafe.patch vendored
View File

@ -1,7 +1,7 @@
From: Marcelo Tosatti <mtosatti@redhat.com>
Date: Wed, 8 Apr 2015 20:33:25 -0300
Subject: KVM: lapic: mark LAPIC timer handler as irqsafe
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Since lapic timer handler only wakes up a simple waitqueue,
it can be executed from hardirq context.

6
debian/patches/features/all/rt/NFSv4-replace-seqcount_t-with-a-seqlock_t.patch vendored
View File

@ -5,7 +5,7 @@ Cc: Anna Schumaker <anna.schumaker@netapp.com>,
linux-nfs@vger.kernel.org, linux-kernel@vger.kernel.org,
tglx@linutronix.de
Subject: NFSv4: replace seqcount_t with a seqlock_t
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The raw_write_seqcount_begin() in nfs4_reclaim_open_state() bugs me
because it maps to preempt_disable() in -RT which I can't have at this
@ -58,7 +58,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2698,7 +2698,7 @@ static int _nfs4_open_and_get_state(stru
@@ -2697,7 +2697,7 @@ static int _nfs4_open_and_get_state(stru
unsigned int seq;
int ret;
@ -67,7 +67,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
ret = _nfs4_proc_open(opendata);
if (ret != 0)
@@ -2736,7 +2736,7 @@ static int _nfs4_open_and_get_state(stru
@@ -2735,7 +2735,7 @@ static int _nfs4_open_and_get_state(stru
if (d_inode(dentry) == state->inode) {
nfs_inode_attach_open_context(ctx);

218
debian/patches/features/all/rt/Revert-timers-Don-t-wake-ktimersoftd-on-every-tick.patch vendored Normal file
View File

@ -0,0 +1,218 @@
From 16145f9c01a2e671aceb731050de9fbf977d31d0 Mon Sep 17 00:00:00 2001
From: Anna-Maria Gleixner <anna-maria@linutronix.de>
Date: Fri, 26 May 2017 19:16:07 +0200
Subject: [PATCH] Revert "timers: Don't wake ktimersoftd on every tick"
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
This reverts commit 032f93cae150a ("timers: Don't wake ktimersoftd on
every tick").
The problem is that the look ahead optimization from the tick timer
interrupt context can race with the softirq thread expiring timer. As
a consequence the temporary hlist heads which hold the to expire
timers are overwritten and the timers which are already removed from
the wheel bucket for expiry are now dangling w/o a list head.
That means those timers never get expired. If one of those timers is
canceled the removal operation will result in a hlist corruption.
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
kernel/time/timer.c | 96 +++++++++++++++-------------------------------------
1 file changed, 29 insertions(+), 67 deletions(-)
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -206,8 +206,6 @@ struct timer_base {
bool is_idle;
DECLARE_BITMAP(pending_map, WHEEL_SIZE);
struct hlist_head vectors[WHEEL_SIZE];
- struct hlist_head expired_lists[LVL_DEPTH];
- int expired_count;
} ____cacheline_aligned;
static DEFINE_PER_CPU(struct timer_base, timer_bases[NR_BASES]);
@@ -1355,8 +1353,7 @@ static void call_timer_fn(struct timer_l
}
}
-static inline void __expire_timers(struct timer_base *base,
- struct hlist_head *head)
+static void expire_timers(struct timer_base *base, struct hlist_head *head)
{
while (!hlist_empty(head)) {
struct timer_list *timer;
@@ -1387,38 +1384,21 @@ static inline void __expire_timers(struc
}
}
-static void expire_timers(struct timer_base *base)
-{
- struct hlist_head *head;
-
- while (base->expired_count--) {
- head = base->expired_lists + base->expired_count;
- __expire_timers(base, head);
- }
- base->expired_count = 0;
-}
-
-static void __collect_expired_timers(struct timer_base *base)
+static int __collect_expired_timers(struct timer_base *base,
+ struct hlist_head *heads)
{
unsigned long clk = base->clk;
struct hlist_head *vec;
- int i;
+ int i, levels = 0;
unsigned int idx;
- /*
- * expire_timers() must be called at least once before we can
- * collect more timers
- */
- if (WARN_ON(base->expired_count))
- return;
-
for (i = 0; i < LVL_DEPTH; i++) {
idx = (clk & LVL_MASK) + i * LVL_SIZE;
if (__test_and_clear_bit(idx, base->pending_map)) {
vec = base->vectors + idx;
- hlist_move_list(vec,
- &base->expired_lists[base->expired_count++]);
+ hlist_move_list(vec, heads++);
+ levels++;
}
/* Is it time to look at the next level? */
if (clk & LVL_CLK_MASK)
@@ -1426,6 +1406,7 @@ static void __collect_expired_timers(str
/* Shift clock for the next level granularity */
clk >>= LVL_CLK_SHIFT;
}
+ return levels;
}
#ifdef CONFIG_NO_HZ_COMMON
@@ -1618,7 +1599,8 @@ void timer_clear_idle(void)
base->is_idle = false;
}
-static void collect_expired_timers(struct timer_base *base)
+static int collect_expired_timers(struct timer_base *base,
+ struct hlist_head *heads)
{
/*
* NOHZ optimization. After a long idle sleep we need to forward the
@@ -1635,49 +1617,20 @@ static void collect_expired_timers(struc
if (time_after(next, jiffies)) {
/* The call site will increment clock! */
base->clk = jiffies - 1;
- return;
+ return 0;
}
base->clk = next;
}
- __collect_expired_timers(base);
+ return __collect_expired_timers(base, heads);
}
#else
-static inline void collect_expired_timers(struct timer_base *base)
+static inline int collect_expired_timers(struct timer_base *base,
+ struct hlist_head *heads)
{
- __collect_expired_timers(base);
+ return __collect_expired_timers(base, heads);
}
#endif
-static int find_expired_timers(struct timer_base *base)
-{
- const unsigned long int end_clk = jiffies;
-
- while (!base->expired_count && time_after_eq(end_clk, base->clk)) {
- collect_expired_timers(base);
- base->clk++;
- }
-
- return base->expired_count;
-}
-
-/* Called from CPU tick routine to quickly collect expired timers */
-static int tick_find_expired(struct timer_base *base)
-{
- int count;
-
- raw_spin_lock(&base->lock);
-
- if (unlikely(time_after(jiffies, base->clk + HZ))) {
- /* defer to ktimersoftd; don't spend too long in irq context */
- count = -1;
- } else
- count = find_expired_timers(base);
-
- raw_spin_unlock(&base->lock);
-
- return count;
-}
-
/*
* Called from the timer interrupt handler to charge one tick to the current
* process. user_tick is 1 if the tick is user time, 0 for system.
@@ -1704,11 +1657,22 @@ void update_process_times(int user_tick)
*/
static inline void __run_timers(struct timer_base *base)
{
+ struct hlist_head heads[LVL_DEPTH];
+ int levels;
+
+ if (!time_after_eq(jiffies, base->clk))
+ return;
+
raw_spin_lock_irq(&base->lock);
- while (find_expired_timers(base))
- expire_timers(base);
+ while (time_after_eq(jiffies, base->clk)) {
+
+ levels = collect_expired_timers(base, heads);
+ base->clk++;
+ while (levels--)
+ expire_timers(base, heads + levels);
+ }
raw_spin_unlock_irq(&base->lock);
wakeup_timer_waiters(base);
}
@@ -1734,12 +1698,12 @@ void run_local_timers(void)
hrtimer_run_queues();
/* Raise the softirq only if required. */
- if (time_before(jiffies, base->clk) || !tick_find_expired(base)) {
+ if (time_before(jiffies, base->clk)) {
if (!IS_ENABLED(CONFIG_NO_HZ_COMMON) || !base->nohz_active)
return;
/* CPU is awake, so check the deferrable base. */
base++;
- if (time_before(jiffies, base->clk) || !tick_find_expired(base))
+ if (time_before(jiffies, base->clk))
return;
}
raise_softirq(TIMER_SOFTIRQ);
@@ -1909,7 +1873,6 @@ int timers_dead_cpu(unsigned int cpu)
raw_spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING);
BUG_ON(old_base->running_timer);
- BUG_ON(old_base->expired_count);
for (i = 0; i < WHEEL_SIZE; i++)
migrate_timer_list(new_base, old_base->vectors + i);
@@ -1936,7 +1899,6 @@ static void __init init_timer_cpu(int cp
#ifdef CONFIG_PREEMPT_RT_FULL
init_swait_queue_head(&base->wait_for_running_timer);
#endif
- base->expired_count = 0;
}
}

2
debian/patches/features/all/rt/acpi-rt-Convert-acpi_gbl_hardware-lock-back-to-a-raw.patch vendored
View File

@ -1,7 +1,7 @@
From: Steven Rostedt <rostedt@goodmis.org>
Date: Wed, 13 Feb 2013 09:26:05 -0500
Subject: acpi/rt: Convert acpi_gbl_hardware lock back to a raw_spinlock_t
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
We hit the following bug with 3.6-rt:

2
debian/patches/features/all/rt/arch-arm64-Add-lazy-preempt-support.patch vendored
View File

@ -1,7 +1,7 @@
From: Anders Roxell <anders.roxell@linaro.org>
Date: Thu, 14 May 2015 17:52:17 +0200
Subject: arch/arm64: Add lazy preempt support
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
arm64 is missing support for PREEMPT_RT. The main feature which is
lacking is support for lazy preemption. The arch-specific entry code,

2
debian/patches/features/all/rt/arm-at91-pit-remove-irq-handler-when-clock-is-unused.patch vendored
View File

@ -1,7 +1,7 @@
From: Benedikt Spranger <b.spranger@linutronix.de>
Date: Sat, 6 Mar 2010 17:47:10 +0100
Subject: ARM: AT91: PIT: Remove irq handler when clock event is unused
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Setup and remove the interrupt handler in clock event mode selection.
This avoids calling the (shared) interrupt handler when the device is

2
debian/patches/features/all/rt/arm-at91-tclib-default-to-tclib-timer-for-rt.patch vendored
View File

@ -1,7 +1,7 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 1 May 2010 18:29:35 +0200
Subject: ARM: at91: tclib: Default to tclib timer for RT
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
RT is not too happy about the shared timer interrupt in AT91
devices. Default to tclib timer for RT.

2
debian/patches/features/all/rt/arm-convert-boot-lock-to-raw.patch vendored
View File

@ -1,7 +1,7 @@
From: Frank Rowand <frank.rowand@am.sony.com>
Date: Mon, 19 Sep 2011 14:51:14 -0700
Subject: arm: Convert arm boot_lock to raw
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The arm boot_lock is used by the secondary processor startup code. The locking
task is the idle thread, which has idle->sched_class == &idle_sched_class.

2
debian/patches/features/all/rt/arm-enable-highmem-for-rt.patch vendored
View File

@ -1,7 +1,7 @@
Subject: arm: Enable highmem for rt
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 13 Feb 2013 11:03:11 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
fixup highmem for ARM.

2
debian/patches/features/all/rt/arm-highmem-flush-tlb-on-unmap.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 11 Mar 2013 21:37:27 +0100
Subject: arm/highmem: Flush tlb on unmap
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The tlb should be flushed on unmap and thus make the mapping entry
invalid. This is only done in the non-debug case which does not look

2
debian/patches/features/all/rt/arm-include-definition-for-cpumask_t.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 22 Dec 2016 17:28:33 +0100
Subject: [PATCH] arm: include definition for cpumask_t
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
This definition gets pulled in by other files. With the (later) split of
RCU and spinlock.h it won't compile anymore.

2
debian/patches/features/all/rt/arm-kprobe-replace-patch_lock-to-raw-lock.patch vendored
View File

@ -1,7 +1,7 @@
From: Yang Shi <yang.shi@linaro.org>
Date: Thu, 10 Nov 2016 16:17:55 -0800
Subject: [PATCH] arm: kprobe: replace patch_lock to raw lock
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
When running kprobe on -rt kernel, the below bug is caught:

2
debian/patches/features/all/rt/arm-preempt-lazy-support.patch vendored
View File

@ -1,7 +1,7 @@
Subject: arm: Add support for lazy preemption
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 31 Oct 2012 12:04:11 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Implement the arm pieces for lazy preempt.

2
debian/patches/features/all/rt/arm-unwind-use_raw_lock.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 20 Sep 2013 14:31:54 +0200
Subject: arm/unwind: use a raw_spin_lock
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Mostly unwind is done with irqs enabled however SLUB may call it with
irqs disabled while creating a new SLUB cache.

2
debian/patches/features/all/rt/arm64-xen--Make-XEN-depend-on-non-rt.patch vendored
View File

@ -1,7 +1,7 @@
Subject: arm64/xen: Make XEN depend on !RT
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 12 Oct 2015 11:18:40 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
It's not ready and probably never will be, unless xen folks have a
look at it.

2
debian/patches/features/all/rt/at91_dont_enable_disable_clock.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Wed, 09 Mar 2016 10:51:06 +0100
Subject: arm: at91: do not disable/enable clocks in a row
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Currently the driver will disable the clock and enable it one line later
if it is switching from periodic mode into one shot.

2
debian/patches/features/all/rt/ata-disable-interrupts-if-non-rt.patch vendored
View File

@ -1,7 +1,7 @@
From: Steven Rostedt <srostedt@redhat.com>
Date: Fri, 3 Jul 2009 08:44:29 -0500
Subject: ata: Do not disable interrupts in ide code for preempt-rt
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Use the local_irq_*_nort variants.

2
debian/patches/features/all/rt/block-blk-mq-use-swait.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 13 Feb 2015 11:01:26 +0100
Subject: block: blk-mq: Use swait
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
| BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:914
| in_atomic(): 1, irqs_disabled(): 0, pid: 255, name: kworker/u257:6

2
debian/patches/features/all/rt/block-mq-don-t-complete-requests-via-IPI.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 29 Jan 2015 15:10:08 +0100
Subject: block/mq: don't complete requests via IPI
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The IPI runs in hardirq context and there are sleeping locks. This patch
moves the completion into a workqueue.

2
debian/patches/features/all/rt/block-mq-drop-preempt-disable.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Tue, 14 Jul 2015 14:26:34 +0200
Subject: block/mq: do not invoke preempt_disable()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
preempt_disable() and get_cpu() don't play well together with the sleeping
locks it tries to allocate later.

2
debian/patches/features/all/rt/block-mq-use-cpu_light.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Wed, 9 Apr 2014 10:37:23 +0200
Subject: block: mq: use cpu_light()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
there is a might sleep splat because get_cpu() disables preemption and
later we grab a lock. As a workaround for this we use get_cpu_light().

12
debian/patches/features/all/rt/block-shorten-interrupt-disabled-regions.patch vendored
View File

@ -1,7 +1,7 @@
Subject: block: Shorten interrupt disabled regions
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 22 Jun 2011 19:47:02 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Moving the blk_sched_flush_plug() call out of the interrupt/preempt
disabled region in the scheduler allows us to replace
@ -48,7 +48,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -3177,7 +3177,7 @@ static void queue_unplugged(struct reque
@@ -3200,7 +3200,7 @@ static void queue_unplugged(struct reque
blk_run_queue_async(q);
else
__blk_run_queue(q);
@ -57,7 +57,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de
}
static void flush_plug_callbacks(struct blk_plug *plug, bool from_schedule)
@@ -3225,7 +3225,6 @@ EXPORT_SYMBOL(blk_check_plugged);
@@ -3248,7 +3248,6 @@ EXPORT_SYMBOL(blk_check_plugged);
void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
{
struct request_queue *q;
@ -65,7 +65,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de
struct request *rq;
LIST_HEAD(list);
unsigned int depth;
@@ -3245,11 +3244,6 @@ void blk_flush_plug_list(struct blk_plug
@@ -3268,11 +3267,6 @@ void blk_flush_plug_list(struct blk_plug
q = NULL;
depth = 0;
@ -77,7 +77,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de
while (!list_empty(&list)) {
rq = list_entry_rq(list.next);
list_del_init(&rq->queuelist);
@@ -3262,7 +3256,7 @@ void blk_flush_plug_list(struct blk_plug
@@ -3285,7 +3279,7 @@ void blk_flush_plug_list(struct blk_plug
queue_unplugged(q, depth, from_schedule);
q = rq->q;
depth = 0;
@ -86,7 +86,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de
}
/*
@@ -3289,8 +3283,6 @@ void blk_flush_plug_list(struct blk_plug
@@ -3312,8 +3306,6 @@ void blk_flush_plug_list(struct blk_plug
*/
if (q)
queue_unplugged(q, depth, from_schedule);

2
debian/patches/features/all/rt/block-use-cpu-chill.patch vendored
View File

@ -1,7 +1,7 @@
Subject: block: Use cpu_chill() for retry loops
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 20 Dec 2012 18:28:26 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Retry loops on RT might loop forever when the modifying side was
preempted. Steven also observed a live lock when there was a

2
debian/patches/features/all/rt/bug-rt-dependend-variants.patch vendored
View File

@ -1,7 +1,7 @@
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:58 -0500
Subject: bug: BUG_ON/WARN_ON variants dependend on RT/!RT
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Introduce RT/NON-RT WARN/BUG statements to avoid ifdefs in the code.

2
debian/patches/features/all/rt/cgroups-scheduling-while-atomic-in-cgroup-code.patch vendored
View File

@ -1,7 +1,7 @@
From: Mike Galbraith <umgwanakikbuti@gmail.com>
Date: Sat, 21 Jun 2014 10:09:48 +0200
Subject: memcontrol: Prevent scheduling while atomic in cgroup code
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
mm, memcg: make refill_stock() use get_cpu_light()

8
debian/patches/features/all/rt/cgroups-use-simple-wait-in-css_release.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 13 Feb 2015 15:52:24 +0100
Subject: cgroups: use simple wait in css_release()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
To avoid:
|BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:914
@ -53,7 +53,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
/*
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -5040,10 +5040,10 @@ static void css_free_rcu_fn(struct rcu_h
@@ -5041,10 +5041,10 @@ static void css_free_rcu_fn(struct rcu_h
queue_work(cgroup_destroy_wq, &css->destroy_work);
}
@ -66,7 +66,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
struct cgroup_subsys *ss = css->ss;
struct cgroup *cgrp = css->cgroup;
@@ -5086,8 +5086,8 @@ static void css_release(struct percpu_re
@@ -5087,8 +5087,8 @@ static void css_release(struct percpu_re
struct cgroup_subsys_state *css =
container_of(ref, struct cgroup_subsys_state, refcnt);
@ -77,7 +77,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
}
static void init_and_link_css(struct cgroup_subsys_state *css,
@@ -5739,6 +5739,7 @@ static int __init cgroup_wq_init(void)
@@ -5740,6 +5740,7 @@ static int __init cgroup_wq_init(void)
*/
cgroup_destroy_wq = alloc_workqueue("cgroup_destroy", 0, 1);
BUG_ON(!cgroup_destroy_wq);

2
debian/patches/features/all/rt/clockevents-drivers-timer-atmel-pit-fix-double-free_.patch vendored
View File

@ -1,7 +1,7 @@
From: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Date: Thu, 17 Mar 2016 21:09:43 +0100
Subject: [PATCH] clockevents/drivers/timer-atmel-pit: fix double free_irq
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
clockevents_exchange_device() changes the state from detached to shutdown
and so at that point the IRQ has not yet been requested.

2
debian/patches/features/all/rt/clocksource-tclib-allow-higher-clockrates.patch vendored
View File

@ -1,7 +1,7 @@
From: Benedikt Spranger <b.spranger@linutronix.de>
Date: Mon, 8 Mar 2010 18:57:04 +0100
Subject: clocksource: TCLIB: Allow higher clock rates for clock events
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
As default the TCLIB uses the 32KiHz base clock rate for clock events.
Add a compile time selection to allow higher clock resulution.

2
debian/patches/features/all/rt/completion-use-simple-wait-queues.patch vendored
View File

@ -1,7 +1,7 @@
Subject: completion: Use simple wait queues
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 11 Jan 2013 11:23:51 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Completions have no long lasting callbacks and therefor do not need
the complex waitqueue variant. Use simple waitqueues which reduces the

2
debian/patches/features/all/rt/cond-resched-lock-rt-tweak.patch vendored
View File

@ -1,7 +1,7 @@
Subject: sched: Use the proper LOCK_OFFSET for cond_resched()
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 22:51:33 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
RT does not increment preempt count when a 'sleeping' spinlock is
locked. Update PREEMPT_LOCK_OFFSET for that case.

8
debian/patches/features/all/rt/cond-resched-softirq-rt.patch vendored
View File

@ -1,7 +1,7 @@
Subject: sched: Take RT softirq semantics into account in cond_resched()
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 14 Jul 2011 09:56:44 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The softirq semantics work different on -RT. There is no SOFTIRQ_MASK in
the preemption counter which leads to the BUG_ON() statement in
@ -16,7 +16,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -3367,12 +3367,16 @@ extern int __cond_resched_lock(spinlock_
@@ -3373,12 +3373,16 @@ extern int __cond_resched_lock(spinlock_
__cond_resched_lock(lock); \
})
@ -35,7 +35,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
{
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5050,6 +5050,7 @@ int __cond_resched_lock(spinlock_t *lock
@@ -5092,6 +5092,7 @@ int __cond_resched_lock(spinlock_t *lock
}
EXPORT_SYMBOL(__cond_resched_lock);
@ -43,7 +43,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
int __sched __cond_resched_softirq(void)
{
BUG_ON(!in_softirq());
@@ -5063,6 +5064,7 @@ int __sched __cond_resched_softirq(void)
@@ -5105,6 +5106,7 @@ int __sched __cond_resched_softirq(void)
return 0;
}
EXPORT_SYMBOL(__cond_resched_softirq);

2
debian/patches/features/all/rt/connector-cn_proc-Protect-send_msg-with-a-local-lock.patch vendored
View File

@ -2,7 +2,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com>
Date: Sun, 16 Oct 2016 05:11:54 +0200
Subject: [PATCH] connector/cn_proc: Protect send_msg() with a local lock
on RT
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
|BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:931
|in_atomic(): 1, irqs_disabled(): 0, pid: 31807, name: sleep

2
debian/patches/features/all/rt/cpu-hotplug-Document-why-PREEMPT_RT-uses-a-spinlock.patch vendored
View File

@ -1,7 +1,7 @@
From: Steven Rostedt <rostedt@goodmis.org>
Date: Thu, 5 Dec 2013 09:16:52 -0500
Subject: cpu hotplug: Document why PREEMPT_RT uses a spinlock
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The patch:

2
debian/patches/features/all/rt/cpu-rt-make-hotplug-lock-a-sleeping-spinlock-on-rt.patch vendored
View File

@ -1,7 +1,7 @@
Subject: cpu: Make hotplug.lock a "sleeping" spinlock on RT
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 02 Mar 2012 10:36:57 -0500
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Tasks can block on hotplug.lock in pin_current_cpu(), but their state
might be != RUNNING. So the mutex wakeup will set the state

6
debian/patches/features/all/rt/cpu-rt-rework-cpu-down.patch vendored
View File

@ -1,7 +1,7 @@
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 16 Jul 2012 08:07:43 +0000
Subject: cpu/rt: Rework cpu down for PREEMPT_RT
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Bringing a CPU down is a pain with the PREEMPT_RT kernel because
tasks can be preempted in many more places than in non-RT. In
@ -57,7 +57,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2474,6 +2474,10 @@ extern void do_set_cpus_allowed(struct t
@@ -2480,6 +2480,10 @@ extern void do_set_cpus_allowed(struct t
extern int set_cpus_allowed_ptr(struct task_struct *p,
const struct cpumask *new_mask);
@ -68,7 +68,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
#else
static inline void do_set_cpus_allowed(struct task_struct *p,
const struct cpumask *new_mask)
@@ -2486,6 +2490,9 @@ static inline int set_cpus_allowed_ptr(s
@@ -2492,6 +2496,9 @@ static inline int set_cpus_allowed_ptr(s
return -EINVAL;
return 0;
}

2
debian/patches/features/all/rt/cpu_chill-Add-a-UNINTERRUPTIBLE-hrtimer_nanosleep.patch vendored
View File

@ -1,7 +1,7 @@
From: Steven Rostedt <rostedt@goodmis.org>
Date: Tue, 4 Mar 2014 12:28:32 -0500
Subject: cpu_chill: Add a UNINTERRUPTIBLE hrtimer_nanosleep
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
We hit another bug that was caused by switching cpu_chill() from
msleep() to hrtimer_nanosleep().

2
debian/patches/features/all/rt/cpu_down_move_migrate_enable_back.patch vendored
View File

@ -1,7 +1,7 @@
From: Tiejun Chen <tiejun.chen@windriver.com>
Subject: cpu_down: move migrate_enable() back
Date: Thu, 7 Nov 2013 10:06:07 +0800
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Commit 08c1ab68, "hotplug-use-migrate-disable.patch", intends to
use migrate_enable()/migrate_disable() to replace that combination

2
debian/patches/features/all/rt/cpufreq-drop-K8-s-driver-from-beeing-selected.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 9 Apr 2015 15:23:01 +0200
Subject: cpufreq: drop K8's driver from beeing selected
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Ralf posted a picture of a backtrace from

2
debian/patches/features/all/rt/cpumask-disable-offstack-on-rt.patch vendored
View File

@ -1,7 +1,7 @@
Subject: cpumask: Disable CONFIG_CPUMASK_OFFSTACK for RT
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 14 Dec 2011 01:03:49 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
There are "valid" GFP_ATOMIC allocations such as

2
debian/patches/features/all/rt/cpuset-Convert-callback_lock-to-raw_spinlock_t.patch vendored
View File

@ -1,7 +1,7 @@
From: Mike Galbraith <efault@gmx.de>
Date: Sun, 8 Jan 2017 09:32:25 +0100
Subject: [PATCH] cpuset: Convert callback_lock to raw_spinlock_t
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The two commits below add up to a cpuset might_sleep() splat for RT:

2
debian/patches/features/all/rt/crypto-Reduce-preempt-disabled-regions-more-algos.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 21 Feb 2014 17:24:04 +0100
Subject: crypto: Reduce preempt disabled regions, more algos
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Don Estabrook reported
| kernel: WARNING: CPU: 2 PID: 858 at kernel/sched/core.c:2428 migrate_disable+0xed/0x100()

2
debian/patches/features/all/rt/debugobjects-rt.patch vendored
View File

@ -1,7 +1,7 @@
Subject: debugobjects: Make RT aware
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 21:41:35 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Avoid filling the pool / allocating memory with irqs off().

2
debian/patches/features/all/rt/dm-make-rt-aware.patch vendored
View File

@ -1,7 +1,7 @@
Subject: dm: Make rt aware
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 14 Nov 2011 23:06:09 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Use the BUG_ON_NORT variant for the irq_disabled() checks. RT has
interrupts legitimately enabled here as we cant deadlock against the

3
debian/patches/features/all/rt/drivers-block-zram-Replace-bit-spinlocks-with-rtmute.patch vendored
View File

@ -2,14 +2,13 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com>
Date: Thu, 31 Mar 2016 04:08:28 +0200
Subject: [PATCH] drivers/block/zram: Replace bit spinlocks with rtmutex
for -rt
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
They're nondeterministic, and lead to ___might_sleep() splats in -rt.
OTOH, they're a lot less wasteful than an rtmutex per page.
Signed-off-by: Mike Galbraith <umgwanakikbuti@gmail.com>
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
[bwh: Adjust context for 4.9.24]
---
drivers/block/zram/zram_drv.c | 30 ++++++++++++++++--------------
drivers/block/zram/zram_drv.h | 41 +++++++++++++++++++++++++++++++++++++++++

2
debian/patches/features/all/rt/drivers-net-8139-disable-irq-nosync.patch vendored
View File

@ -1,7 +1,7 @@
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:24 -0500
Subject: drivers/net: Use disable_irq_nosync() in 8139too
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Use disable_irq_nosync() instead of disable_irq() as this might be
called in atomic context with netpoll.

2
debian/patches/features/all/rt/drivers-net-vortex-fix-locking-issues.patch vendored
View File

@ -1,7 +1,7 @@
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 3 Jul 2009 08:30:00 -0500
Subject: drivers/net: vortex fix locking issues
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Argh, cut and paste wasn't enough...

2
debian/patches/features/all/rt/drivers-random-reduce-preempt-disabled-region.patch vendored
View File

@ -1,7 +1,7 @@
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:30 -0500
Subject: drivers: random: Reduce preempt disabled region
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
No need to keep preemption disabled across the whole function.

2
debian/patches/features/all/rt/drivers-tty-fix-omap-lock-crap.patch vendored
View File

@ -1,7 +1,7 @@
Subject: tty/serial/omap: Make the locking RT aware
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 28 Jul 2011 13:32:57 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The lock is a sleeping lock and local_irq_save() is not the
optimsation we are looking for. Redo it to make it work on -RT and

2
debian/patches/features/all/rt/drivers-tty-pl011-irq-disable-madness.patch vendored
View File

@ -1,7 +1,7 @@
Subject: tty/serial/pl011: Make the locking work on RT
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 08 Jan 2013 21:36:51 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The lock is a sleeping lock and local_irq_save() is not the optimsation
we are looking for. Redo it to make it work on -RT and non-RT.

3
debian/patches/features/all/rt/drivers-zram-Don-t-disable-preemption-in-zcomp_strea.patch vendored
View File

@ -2,7 +2,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com>
Date: Thu, 20 Oct 2016 11:15:22 +0200
Subject: [PATCH] drivers/zram: Don't disable preemption in
zcomp_stream_get/put()
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
In v4.7, the driver switched to percpu compression streams, disabling
preemption via get/put_cpu_ptr(). Use a per-zcomp_strm lock here. We
@ -13,7 +13,6 @@ zram_bvec_write().
Signed-off-by: Mike Galbraith <umgwanakikbuti@gmail.com>
[bigeasy: get_locked_var() -> per zcomp_strm lock]
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
[bwh: Adjust context for 4.9.24]
---
drivers/block/zram/zcomp.c | 12 ++++++++++--
drivers/block/zram/zcomp.h | 1 +

4
debian/patches/features/all/rt/drm-i915-drop-trace_i915_gem_ring_dispatch-onrt.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 25 Apr 2013 18:12:52 +0200
Subject: drm/i915: drop trace_i915_gem_ring_dispatch on rt
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
This tracepoint is responsible for:
@ -47,7 +47,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1537,7 +1537,9 @@ execbuf_submit(struct i915_execbuffer_pa
@@ -1489,7 +1489,9 @@ execbuf_submit(struct i915_execbuffer_pa
if (ret)
return ret;

2
debian/patches/features/all/rt/drmi915_Use_local_lockunlock_irq()_in_intel_pipe_update_startend().patch vendored
View File

@ -1,7 +1,7 @@
Subject: drm,i915: Use local_lock/unlock_irq() in intel_pipe_update_start/end()
From: Mike Galbraith <umgwanakikbuti@gmail.com>
Date: Sat, 27 Feb 2016 09:01:42 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
[ 8.014039] BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:918

2
debian/patches/features/all/rt/drmradeoni915_Use_preempt_disableenable_rt()_where_recommended.patch vendored
View File

@ -1,7 +1,7 @@
Subject: drm,radeon,i915: Use preempt_disable/enable_rt() where recommended
From: Mike Galbraith <umgwanakikbuti@gmail.com>
Date: Sat, 27 Feb 2016 08:09:11 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
DRM folks identified the spots, so use them.

2
debian/patches/features/all/rt/epoll-use-get-cpu-light.patch vendored
View File

@ -1,7 +1,7 @@
Subject: fs/epoll: Do not disable preemption on RT
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 08 Jul 2011 16:35:35 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
ep_call_nested() takes a sleeping lock so we can't disable preemption.
The light version is enough since ep_call_nested() doesn't mind beeing

2
debian/patches/features/all/rt/fs-aio-simple-simple-work.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 16 Feb 2015 18:49:10 +0100
Subject: fs/aio: simple simple work
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
|BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:768
|in_atomic(): 1, irqs_disabled(): 0, pid: 26, name: rcuos/2

2
debian/patches/features/all/rt/fs-block-rt-support.patch vendored
View File

@ -1,7 +1,7 @@
Subject: block: Turn off warning which is bogus on RT
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 14 Jun 2011 17:05:09 +0200
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
On -RT the context is always with IRQs enabled. Ignore this warning on -RT.

2
debian/patches/features/all/rt/fs-dcache-include-wait.h.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Wed, 14 Sep 2016 11:55:23 +0200
Subject: fs/dcache: include wait.h
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Since commit d9171b934526 ("parallel lookups machinery, part 4 (and
last)") dcache.h is using but does not include wait.h. It works as long

2
debian/patches/features/all/rt/fs-dcache-init-in_lookup_hashtable.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Wed, 14 Sep 2016 17:57:03 +0200
Subject: [PATCH] fs/dcache: init in_lookup_hashtable
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
in_lookup_hashtable was introduced in commit 94bdd655caba ("parallel
lookups machinery, part 3") and never initialized but since it is in

2
debian/patches/features/all/rt/fs-dcache-use-cpu-chill-in-trylock-loops.patch vendored
View File

@ -1,7 +1,7 @@
Subject: fs: dcache: Use cpu_chill() in trylock loops
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 07 Mar 2012 21:00:34 +0100
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Retry loops on RT might loop forever when the modifying side was
preempted. Use cpu_chill() instead of cpu_relax() to let the system

2
debian/patches/features/all/rt/fs-dcache-use-swait_queue-instead-of-waitqueue.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Wed, 14 Sep 2016 14:35:49 +0200
Subject: [PATCH] fs/dcache: use swait_queue instead of waitqueue
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
__d_lookup_done() invokes wake_up_all() while holding a hlist_bl_lock()
which disables preemption. As a workaround convert it to swait.

2
debian/patches/features/all/rt/fs-jbd-replace-bh_state-lock.patch vendored
View File

@ -1,7 +1,7 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 18 Mar 2011 10:11:25 +0100
Subject: fs: jbd/jbd2: Make state lock and journal head lock rt safe
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
bit_spin_locks break under RT.

2
debian/patches/features/all/rt/fs-jbd2-pull-your-plug-when-waiting-for-space.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Mon, 17 Feb 2014 17:30:03 +0100
Subject: fs: jbd2: pull your plug when waiting for space
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Two cps in parallel managed to stall the the ext4 fs. It seems that
journal code is either waiting for locks or sleeping waiting for

2
debian/patches/features/all/rt/fs-namespace-preemption-fix.patch vendored
View File

@ -1,7 +1,7 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 19 Jul 2009 08:44:27 -0500
Subject: fs: namespace preemption fix
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
On RT we cannot loop with preemption disabled here as
mnt_make_readonly() might have been preempted. We can safely enable

2
debian/patches/features/all/rt/fs-nfs-turn-rmdir_sem-into-a-semaphore.patch vendored
View File

@ -1,7 +1,7 @@
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Thu, 15 Sep 2016 10:51:27 +0200
Subject: [PATCH] fs/nfs: turn rmdir_sem into a semaphore
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
The RW semaphore had a reader side which used the _non_owner version
because it most likely took the reader lock in one thread and released it

2
debian/patches/features/all/rt/fs-ntfs-disable-interrupt-non-rt.patch vendored
View File

@ -1,7 +1,7 @@
From: Mike Galbraith <efault@gmx.de>
Date: Fri, 3 Jul 2009 08:44:12 -0500
Subject: fs: ntfs: disable interrupt only on !RT
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
On Sat, 2007-10-27 at 11:44 +0200, Ingo Molnar wrote:
> * Nick Piggin <nickpiggin@yahoo.com.au> wrote:

2
debian/patches/features/all/rt/fs-replace-bh_uptodate_lock-for-rt.patch vendored
View File

@ -1,7 +1,7 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 18 Mar 2011 09:18:52 +0100
Subject: buffer_head: Replace bh_uptodate_lock for -rt
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Wrap the bit_spin_lock calls into a separate inline and add the RT
replacements with a real spinlock.

2
debian/patches/features/all/rt/ftrace-Fix-trace-header-alignment.patch vendored
View File

@ -1,7 +1,7 @@
From: Mike Galbraith <umgwanakikbuti@gmail.com>
Date: Sun, 16 Oct 2016 05:08:30 +0200
Subject: [PATCH] ftrace: Fix trace header alignment
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Line up helper arrows to the right column.

2
debian/patches/features/all/rt/ftrace-migrate-disable-tracing.patch vendored
View File

@ -1,7 +1,7 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 21:56:42 +0200
Subject: trace: Add migrate-disabled counter to tracing output
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---

2
debian/patches/features/all/rt/futex-Ensure-lock-unlock-symetry-versus-pi_lock-and-.patch vendored
View File

@ -1,7 +1,7 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 1 Mar 2013 11:17:42 +0100
Subject: futex: Ensure lock/unlock symetry versus pi_lock and hash bucket lock
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
In exit_pi_state_list() we have the following locking construct:

6
debian/patches/features/all/rt/futex-requeue-pi-fix.patch vendored
View File

@ -1,7 +1,7 @@
From: Steven Rostedt <rostedt@goodmis.org>
Date: Tue, 14 Jul 2015 14:26:34 +0200
Subject: futex: Fix bug on when a requeued RT task times out
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.20-rt16.tar.xz
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.9/older/patches-4.9.30-rt20.tar.xz
Requeue with timeout causes a bug with PREEMPT_RT_FULL.
@ -66,7 +66,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
}
/*
@@ -1696,6 +1697,35 @@ int __rt_mutex_start_proxy_lock(struct r
@@ -1712,6 +1713,35 @@ int __rt_mutex_start_proxy_lock(struct r
if (try_to_take_rt_mutex(lock, task, NULL))
return 1;
@ -104,7 +104,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
RT_MUTEX_FULL_CHAINWALK);
--- a/kernel/locking/rtmutex_common.h
+++ b/kernel/locking/rtmutex_common.h
@@ -99,6 +99,7 @@ enum rtmutex_chainwalk {
@@ -100,6 +100,7 @@ enum rtmutex_chainwalk {
* PI-futex support (proxy locking functions, etc.):
*/
#define PI_WAKEUP_INPROGRESS ((struct rt_mutex_waiter *) 1)

Some files were not shown because too many files have changed in this diff Show More