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

[rt] Update to 3.4-rt8 and reenable 

svn path=/dists/trunk/linux-2.6/; revision=19085
This commit is contained in:
Ben Hutchings 2012-06-04 21:20:09 +00:00
parent d866f5e288 1e9083ce67
commit 478d361442
300 changed files with 6308 additions and 8565 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
debian/changelog vendored
View File

@ -13,6 +13,7 @@ linux-2.6 (3.4.1-1~experimental.1) UNRELEASED; urgency=low
[ Ben Hutchings ]
* DFSG: Remove the new vs6624 driver, which contains non-free firmware
* aufs: Update to aufs3.4-20120521
* [rt] Update to 3.4-rt8 and reenable
-- Ben Hutchings <ben@decadent.org.uk> Mon, 21 May 2012 01:37:21 +0100

2
debian/config/amd64/defines vendored
View File

@ -1,7 +1,7 @@
[base]
featuresets:
none
# rt
rt
kernel-arch: x86
[build]

2
debian/config/i386/defines vendored
View File

@ -1,7 +1,7 @@
[base]
featuresets:
none
# rt
rt
kernel-arch: x86
[description]

155
debian/patches/features/all/rt/0001-x86-Call-idle-notifier-after-irq_enter.patch vendored
View File

@ -1,155 +0,0 @@
From 76f41af7cc93591a2723d2c59918df003f90cb09 Mon Sep 17 00:00:00 2001
From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Mon, 26 Sep 2011 12:19:11 +0200
Subject: [PATCH 001/267] x86: Call idle notifier after irq_enter()
Interrupts notify the idle exit state before calling irq_enter(). But
the notifier code calls rcu_read_lock() and this is not allowed while
rcu is in an extended quiescent state. We need to wait for
rcu_irq_enter() to be called before doing so otherwise this results in
a grumpy RCU:
[ 0.099991] WARNING: at include/linux/rcupdate.h:194 __atomic_notifier_call_chain+0xd2/0x110()
[ 0.099991] Hardware name: AMD690VM-FMH
[ 0.099991] Modules linked in:
[ 0.099991] Pid: 0, comm: swapper Not tainted 3.0.0-rc6+ #255
[ 0.099991] Call Trace:
[ 0.099991] <IRQ> [<ffffffff81051c8a>] warn_slowpath_common+0x7a/0xb0
[ 0.099991] [<ffffffff81051cd5>] warn_slowpath_null+0x15/0x20
[ 0.099991] [<ffffffff817d6fa2>] __atomic_notifier_call_chain+0xd2/0x110
[ 0.099991] [<ffffffff817d6ff1>] atomic_notifier_call_chain+0x11/0x20
[ 0.099991] [<ffffffff81001873>] exit_idle+0x43/0x50
[ 0.099991] [<ffffffff81020439>] smp_apic_timer_interrupt+0x39/0xa0
[ 0.099991] [<ffffffff817da253>] apic_timer_interrupt+0x13/0x20
[ 0.099991] <EOI> [<ffffffff8100ae67>] ? default_idle+0xa7/0x350
[ 0.099991] [<ffffffff8100ae65>] ? default_idle+0xa5/0x350
[ 0.099991] [<ffffffff8100b19b>] amd_e400_idle+0x8b/0x110
[ 0.099991] [<ffffffff810cb01f>] ? rcu_enter_nohz+0x8f/0x160
[ 0.099991] [<ffffffff810019a0>] cpu_idle+0xb0/0x110
[ 0.099991] [<ffffffff817a7505>] rest_init+0xe5/0x140
[ 0.099991] [<ffffffff817a7468>] ? rest_init+0x48/0x140
[ 0.099991] [<ffffffff81cc5ca3>] start_kernel+0x3d1/0x3dc
[ 0.099991] [<ffffffff81cc5321>] x86_64_start_reservations+0x131/0x135
[ 0.099991] [<ffffffff81cc5412>] x86_64_start_kernel+0xed/0xf4
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Link: http://lkml.kernel.org/r/20110929194047.GA10247@linux.vnet.ibm.com
Cc: Ingo Molnar <mingo@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Andy Henroid <andrew.d.henroid@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/kernel/apic/apic.c | 6 +++---
arch/x86/kernel/apic/io_apic.c | 2 +-
arch/x86/kernel/cpu/mcheck/therm_throt.c | 2 +-
arch/x86/kernel/cpu/mcheck/threshold.c | 2 +-
arch/x86/kernel/irq.c | 6 +++---
5 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c
index f98d84c..2cd2d93 100644
--- a/arch/x86/kernel/apic/apic.c
+++ b/arch/x86/kernel/apic/apic.c
@@ -876,8 +876,8 @@ void __irq_entry smp_apic_timer_interrupt(struct pt_regs *regs)
* Besides, if we don't timer interrupts ignore the global
* interrupt lock, which is the WrongThing (tm) to do.
*/
- exit_idle();
irq_enter();
+ exit_idle();
local_apic_timer_interrupt();
irq_exit();
@@ -1809,8 +1809,8 @@ void smp_spurious_interrupt(struct pt_regs *regs)
{
u32 v;
- exit_idle();
irq_enter();
+ exit_idle();
/*
* Check if this really is a spurious interrupt and ACK it
* if it is a vectored one. Just in case...
@@ -1846,8 +1846,8 @@ void smp_error_interrupt(struct pt_regs *regs)
"Illegal register address", /* APIC Error Bit 7 */
};
- exit_idle();
irq_enter();
+ exit_idle();
/* First tickle the hardware, only then report what went on. -- REW */
v0 = apic_read(APIC_ESR);
apic_write(APIC_ESR, 0);
diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
index 6d939d7..8980555 100644
--- a/arch/x86/kernel/apic/io_apic.c
+++ b/arch/x86/kernel/apic/io_apic.c
@@ -2421,8 +2421,8 @@ asmlinkage void smp_irq_move_cleanup_interrupt(void)
unsigned vector, me;
ack_APIC_irq();
- exit_idle();
irq_enter();
+ exit_idle();
me = smp_processor_id();
for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) {
diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c
index 787e06c..ce21561 100644
--- a/arch/x86/kernel/cpu/mcheck/therm_throt.c
+++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c
@@ -397,8 +397,8 @@ static void (*smp_thermal_vector)(void) = unexpected_thermal_interrupt;
asmlinkage void smp_thermal_interrupt(struct pt_regs *regs)
{
- exit_idle();
irq_enter();
+ exit_idle();
inc_irq_stat(irq_thermal_count);
smp_thermal_vector();
irq_exit();
diff --git a/arch/x86/kernel/cpu/mcheck/threshold.c b/arch/x86/kernel/cpu/mcheck/threshold.c
index d746df2..aa578ca 100644
--- a/arch/x86/kernel/cpu/mcheck/threshold.c
+++ b/arch/x86/kernel/cpu/mcheck/threshold.c
@@ -19,8 +19,8 @@ void (*mce_threshold_vector)(void) = default_threshold_interrupt;
asmlinkage void smp_threshold_interrupt(void)
{
- exit_idle();
irq_enter();
+ exit_idle();
inc_irq_stat(irq_threshold_count);
mce_threshold_vector();
irq_exit();
diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c
index 429e0c9..5d31e5b 100644
--- a/arch/x86/kernel/irq.c
+++ b/arch/x86/kernel/irq.c
@@ -181,8 +181,8 @@ unsigned int __irq_entry do_IRQ(struct pt_regs *regs)
unsigned vector = ~regs->orig_ax;
unsigned irq;
- exit_idle();
irq_enter();
+ exit_idle();
irq = __this_cpu_read(vector_irq[vector]);
@@ -209,10 +209,10 @@ void smp_x86_platform_ipi(struct pt_regs *regs)
ack_APIC_irq();
- exit_idle();
-
irq_enter();
+ exit_idle();
+
inc_irq_stat(x86_platform_ipis);
if (x86_platform_ipi_callback)
--
1.7.10

27
debian/patches/features/all/rt/0008-arm-enable-interrupts-in-signal-code.patch.patch vendored
View File

@ -1,27 +0,0 @@
From 415c7e81f6b54e5f8e8cddf0f2ff1c1855eeade3 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 16 Jul 2011 16:27:13 +0200
Subject: [PATCH 008/267] arm-enable-interrupts-in-signal-code.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/kernel/signal.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c
index 9e617bd..c7001bc 100644
--- a/arch/arm/kernel/signal.c
+++ b/arch/arm/kernel/signal.c
@@ -672,6 +672,9 @@ static void do_signal(struct pt_regs *regs, int syscall)
if (!user_mode(regs))
return;
+ local_irq_enable();
+ preempt_check_resched();
+
/*
* If we were from a system call, check for system call restarting...
*/
--
1.7.10

28
debian/patches/features/all/rt/0009-powerpc-85xx-Mark-cascade-irq-IRQF_NO_THREAD.patch vendored
View File

@ -1,28 +0,0 @@
From 490faa4e4541df017386a562e4e9ad7011f49562 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 16 Jul 2011 12:09:54 +0200
Subject: [PATCH 009/267] powerpc: 85xx: Mark cascade irq IRQF_NO_THREAD
Cascade interrupt must run in hard interrupt context.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/powerpc/platforms/85xx/mpc85xx_cds.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/platforms/85xx/mpc85xx_cds.c b/arch/powerpc/platforms/85xx/mpc85xx_cds.c
index 66cb8d6..7e2a4d2 100644
--- a/arch/powerpc/platforms/85xx/mpc85xx_cds.c
+++ b/arch/powerpc/platforms/85xx/mpc85xx_cds.c
@@ -177,7 +177,7 @@ static irqreturn_t mpc85xx_8259_cascade_action(int irq, void *dev_id)
static struct irqaction mpc85xxcds_8259_irqaction = {
.handler = mpc85xx_8259_cascade_action,
- .flags = IRQF_SHARED,
+ .flags = IRQF_SHARED | IRQF_NO_THREAD,
.name = "8259 cascade",
};
#endif /* PPC_I8259 */
--
1.7.10

30
debian/patches/features/all/rt/0010-powerpc-wsp-Mark-opb-cascade-handler-IRQF_NO_THREAD.patch vendored
View File

@ -1,30 +0,0 @@
From 05a48c448ba9085941f020ea02a7c880e80d683c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 5 Oct 2011 14:11:24 +0200
Subject: [PATCH 010/267] powerpc: wsp: Mark opb cascade handler
IRQF_NO_THREAD
Cascade handlers must run in hard interrupt context.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/powerpc/platforms/wsp/opb_pic.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/platforms/wsp/opb_pic.c b/arch/powerpc/platforms/wsp/opb_pic.c
index be05631..19f353d 100644
--- a/arch/powerpc/platforms/wsp/opb_pic.c
+++ b/arch/powerpc/platforms/wsp/opb_pic.c
@@ -320,7 +320,8 @@ void __init opb_pic_init(void)
}
/* Attach opb interrupt handler to new virtual IRQ */
- rc = request_irq(virq, opb_irq_handler, 0, "OPB LS Cascade", opb);
+ rc = request_irq(virq, opb_irq_handler, IRQF_NO_THREAD,
+ "OPB LS Cascade", opb);
if (rc) {
printk("opb: request_irq failed: %d\n", rc);
continue;
--
1.7.10

73
debian/patches/features/all/rt/0011-powerpc-Mark-IPI-interrupts-IRQF_NO_THREAD.patch vendored
View File

@ -1,73 +0,0 @@
From c5f5de2dcb79c63ab914f69b8d1a59ba211f8b15 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 5 Oct 2011 14:00:26 +0200
Subject: [PATCH 011/267] powerpc: Mark IPI interrupts IRQF_NO_THREAD
IPI handlers cannot be threaded. Remove the obsolete IRQF_DISABLED
flag (see commit e58aa3d2) while at it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/powerpc/kernel/smp.c | 4 ++--
arch/powerpc/platforms/powermac/smp.c | 4 ++--
arch/powerpc/sysdev/xics/xics-common.c | 5 +++--
3 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 6df7090..abdedd3 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -187,8 +187,8 @@ int smp_request_message_ipi(int virq, int msg)
return 1;
}
#endif
- err = request_irq(virq, smp_ipi_action[msg], IRQF_PERCPU,
- smp_ipi_name[msg], 0);
+ err = request_irq(virq, smp_ipi_action[msg],
+ IRQF_PERCPU | IRQF_NO_THREAD, smp_ipi_name[msg], 0);
WARN(err < 0, "unable to request_irq %d for %s (rc %d)\n",
virq, smp_ipi_name[msg], err);
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index 3394254..8d75ac8 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -200,7 +200,7 @@ static int psurge_secondary_ipi_init(void)
if (psurge_secondary_virq)
rc = request_irq(psurge_secondary_virq, psurge_ipi_intr,
- IRQF_PERCPU, "IPI", NULL);
+ IRQF_NO_THREAD | IRQF_PERCPU, "IPI", NULL);
if (rc)
pr_err("Failed to setup secondary cpu IPI\n");
@@ -408,7 +408,7 @@ static int __init smp_psurge_kick_cpu(int nr)
static struct irqaction psurge_irqaction = {
.handler = psurge_ipi_intr,
- .flags = IRQF_PERCPU,
+ .flags = IRQF_PERCPU | IRQF_NO_THREAD,
.name = "primary IPI",
};
diff --git a/arch/powerpc/sysdev/xics/xics-common.c b/arch/powerpc/sysdev/xics/xics-common.c
index 63762c6..4ba6194 100644
--- a/arch/powerpc/sysdev/xics/xics-common.c
+++ b/arch/powerpc/sysdev/xics/xics-common.c
@@ -134,10 +134,11 @@ static void xics_request_ipi(void)
BUG_ON(ipi == NO_IRQ);
/*
- * IPIs are marked IRQF_PERCPU. The handler was set in map.
+ * IPIs are marked PERCPU and also IRQF_NO_THREAD as they must
+ * run in hard interrupt context. The handler was set in map.
*/
BUG_ON(request_irq(ipi, icp_ops->ipi_action,
- IRQF_PERCPU, "IPI", NULL));
+ IRQF_NO_THREAD|IRQF_PERCPU, "IPI", NULL));
}
int __init xics_smp_probe(void)
--
1.7.10

28
debian/patches/features/all/rt/0012-powerpc-Allow-irq-threading.patch vendored
View File

@ -1,28 +0,0 @@
From 2e574f68fff54d8c7398b3984f183fb80ec8ce3e Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 16 Jul 2011 13:16:24 +0200
Subject: [PATCH 012/267] powerpc: Allow irq threading
All interrupts which must be non threaded are marked
IRQF_NO_THREAD. So it's safe to allow force threaded handlers.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/powerpc/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 951e18f..7c93c7e 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -132,6 +132,7 @@ config PPC
select IRQ_PER_CPU
select GENERIC_IRQ_SHOW
select GENERIC_IRQ_SHOW_LEVEL
+ select IRQ_FORCED_THREADING
select HAVE_RCU_TABLE_FREE if SMP
select HAVE_SYSCALL_TRACEPOINTS
select HAVE_BPF_JIT if (PPC64 && NET)
--
1.7.10

68
debian/patches/features/all/rt/0013-sched-Keep-period-timer-ticking-when-throttling-acti.patch vendored
View File

@ -1,68 +0,0 @@
From 9e1c03d11fdc2426a5d53a1d0e0b63b412924287 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Tue, 18 Oct 2011 22:03:48 +0200
Subject: [PATCH 013/267] sched: Keep period timer ticking when throttling
active
When a runqueue is throttled we cannot disable the period timer
because that timer is the only way to undo the throttling.
We got stale throttling entries when a rq was throttled and then the
global sysctl was disabled, which stopped the timer.
[ tglx: Preliminary changelog ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched_rt.c | 13 ++++++++-----
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 78fcacf..40d97e1 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -580,12 +580,9 @@ static inline int balance_runtime(struct rt_rq *rt_rq)
static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
{
- int i, idle = 1;
+ int i, idle = 1, throttled = 0;
const struct cpumask *span;
- if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF)
- return 1;
-
span = sched_rt_period_mask();
for_each_cpu(i, span) {
int enqueue = 0;
@@ -620,12 +617,17 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
if (!rt_rq_throttled(rt_rq))
enqueue = 1;
}
+ if (rt_rq->rt_throttled)
+ throttled = 1;
if (enqueue)
sched_rt_rq_enqueue(rt_rq);
raw_spin_unlock(&rq->lock);
}
+ if (!throttled && (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF))
+ return 1;
+
return idle;
}
@@ -686,7 +688,8 @@ static void update_curr_rt(struct rq *rq)
if (unlikely((s64)delta_exec < 0))
delta_exec = 0;
- schedstat_set(curr->se.statistics.exec_max, max(curr->se.statistics.exec_max, delta_exec));
+ schedstat_set(curr->se.statistics.exec_max,
+ max(curr->se.statistics.exec_max, delta_exec));
curr->se.sum_exec_runtime += delta_exec;
account_group_exec_runtime(curr, delta_exec);
--
1.7.10

52
debian/patches/features/all/rt/0014-sched-Do-not-throttle-due-to-PI-boosting.patch vendored
View File

@ -1,52 +0,0 @@
From ce43773df1fea5d063ac8d1aca4ff806bf899b5b Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Tue, 18 Oct 2011 22:03:48 +0200
Subject: [PATCH 014/267] sched: Do not throttle due to PI boosting
When a runqueue has rt_runtime_us = 0 then the only way it can
accumulate rt_time is via PI boosting. Though that causes the runqueue
to be throttled and replenishing does not change anything due to
rt_runtime_us = 0. So avoid that situation by clearing rt_time and
skip the throttling alltogether.
[ tglx: Preliminary changelog ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched_rt.c | 20 ++++++++++++++++++--
1 file changed, 18 insertions(+), 2 deletions(-)
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 40d97e1..c108b9c 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -659,8 +659,24 @@ static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
return 0;
if (rt_rq->rt_time > runtime) {
- rt_rq->rt_throttled = 1;
- printk_once(KERN_WARNING "sched: RT throttling activated\n");
+ struct rt_bandwidth *rt_b = sched_rt_bandwidth(rt_rq);
+
+ /*
+ * Don't actually throttle groups that have no runtime assigned
+ * but accrue some time due to boosting.
+ */
+ if (likely(rt_b->rt_runtime)) {
+ rt_rq->rt_throttled = 1;
+ printk_once(KERN_WARNING "sched: RT throttling activated\n");
+ } else {
+ /*
+ * In case we did anyway, make it go away,
+ * replenishment is a joke, since it will replenish us
+ * with exactly 0 ns.
+ */
+ rt_rq->rt_time = 0;
+ }
+
if (rt_rq_throttled(rt_rq)) {
sched_rt_rq_dequeue(rt_rq);
return 1;
--
1.7.10

42
debian/patches/features/all/rt/0015-time-Remove-bogus-comments.patch vendored
View File

@ -1,42 +0,0 @@
From f427d80501aa7fcdf695288aa8eb507dbd5c9369 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Feb 2012 19:06:50 +0100
Subject: [PATCH 015/267] time: Remove bogus comments
There is no global irq lock which makes a syscall magically SMP
safe. Remove the outdated comment concerning do_settimeofday() as
well.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/time.c | 6 ------
1 file changed, 6 deletions(-)
diff --git a/kernel/time.c b/kernel/time.c
index 73e416d..ba744cf 100644
--- a/kernel/time.c
+++ b/kernel/time.c
@@ -163,7 +163,6 @@ int do_sys_settimeofday(const struct timespec *tv, const struct timezone *tz)
return error;
if (tz) {
- /* SMP safe, global irq locking makes it work. */
sys_tz = *tz;
update_vsyscall_tz();
if (firsttime) {
@@ -173,12 +172,7 @@ int do_sys_settimeofday(const struct timespec *tv, const struct timezone *tz)
}
}
if (tv)
- {
- /* SMP safe, again the code in arch/foo/time.c should
- * globally block out interrupts when it runs.
- */
return do_settimeofday(tv);
- }
return 0;
}
--
1.7.10

37
debian/patches/features/all/rt/0016-x86-vdso-Remove-bogus-locking-in-update_vsyscall_tz.patch vendored
View File

@ -1,37 +0,0 @@
From 76f63af187b74e8d533678278b247292340ccc10 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Feb 2012 19:10:46 +0100
Subject: [PATCH 016/267] x86: vdso: Remove bogus locking in
update_vsyscall_tz()
Changing the sequence count in update_vsyscall_tz() is completely
pointless.
The vdso code copies the data unprotected. There is no point to change
this as sys_tz is nowhere protected at all. See sys_gettimeofday().
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/kernel/vsyscall_64.c | 5 -----
1 file changed, 5 deletions(-)
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index e4d4a22..f04adbd 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -80,12 +80,7 @@ early_param("vsyscall", vsyscall_setup);
void update_vsyscall_tz(void)
{
- unsigned long flags;
-
- write_seqlock_irqsave(&vsyscall_gtod_data.lock, flags);
- /* sys_tz has changed */
vsyscall_gtod_data.sys_tz = sys_tz;
- write_sequnlock_irqrestore(&vsyscall_gtod_data.lock, flags);
}
void update_vsyscall(struct timespec *wall_time, struct timespec *wtm,
--
1.7.10

128
debian/patches/features/all/rt/0017-x86-vdso-Use-seqcount-instead-of-seqlock.patch vendored
View File

@ -1,128 +0,0 @@
From 11f9d4b70b2592b57dda3c33bdcd726ee98c9a71 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Feb 2012 18:24:07 +0100
Subject: [PATCH 017/267] x86: vdso: Use seqcount instead of seqlock
The update of the vdso data happens under xtime_lock, so adding a
nested lock is pointless. Just use a seqcount to sync the readers.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/include/asm/vgtod.h | 2 +-
arch/x86/kernel/vsyscall_64.c | 11 +++--------
arch/x86/vdso/vclock_gettime.c | 16 ++++++++--------
3 files changed, 12 insertions(+), 17 deletions(-)
diff --git a/arch/x86/include/asm/vgtod.h b/arch/x86/include/asm/vgtod.h
index 815285b..1f00717 100644
--- a/arch/x86/include/asm/vgtod.h
+++ b/arch/x86/include/asm/vgtod.h
@@ -5,7 +5,7 @@
#include <linux/clocksource.h>
struct vsyscall_gtod_data {
- seqlock_t lock;
+ seqcount_t seq;
/* open coded 'struct timespec' */
time_t wall_time_sec;
diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c
index f04adbd..50392ee 100644
--- a/arch/x86/kernel/vsyscall_64.c
+++ b/arch/x86/kernel/vsyscall_64.c
@@ -52,10 +52,7 @@
#include "vsyscall_trace.h"
DEFINE_VVAR(int, vgetcpu_mode);
-DEFINE_VVAR(struct vsyscall_gtod_data, vsyscall_gtod_data) =
-{
- .lock = __SEQLOCK_UNLOCKED(__vsyscall_gtod_data.lock),
-};
+DEFINE_VVAR(struct vsyscall_gtod_data, vsyscall_gtod_data);
static enum { EMULATE, NATIVE, NONE } vsyscall_mode = NATIVE;
@@ -86,9 +83,7 @@ void update_vsyscall_tz(void)
void update_vsyscall(struct timespec *wall_time, struct timespec *wtm,
struct clocksource *clock, u32 mult)
{
- unsigned long flags;
-
- write_seqlock_irqsave(&vsyscall_gtod_data.lock, flags);
+ write_seqcount_begin(&vsyscall_gtod_data.seq);
/* copy vsyscall data */
vsyscall_gtod_data.clock.vclock_mode = clock->archdata.vclock_mode;
@@ -101,7 +96,7 @@ void update_vsyscall(struct timespec *wall_time, struct timespec *wtm,
vsyscall_gtod_data.wall_to_monotonic = *wtm;
vsyscall_gtod_data.wall_time_coarse = __current_kernel_time();
- write_sequnlock_irqrestore(&vsyscall_gtod_data.lock, flags);
+ write_seqcount_end(&vsyscall_gtod_data.seq);
}
static void warn_bad_vsyscall(const char *level, struct pt_regs *regs,
diff --git a/arch/x86/vdso/vclock_gettime.c b/arch/x86/vdso/vclock_gettime.c
index 6bc0e72..d8511fb 100644
--- a/arch/x86/vdso/vclock_gettime.c
+++ b/arch/x86/vdso/vclock_gettime.c
@@ -86,11 +86,11 @@ notrace static noinline int do_realtime(struct timespec *ts)
{
unsigned long seq, ns;
do {
- seq = read_seqbegin(&gtod->lock);
+ seq = read_seqcount_begin(&gtod->seq);
ts->tv_sec = gtod->wall_time_sec;
ts->tv_nsec = gtod->wall_time_nsec;
ns = vgetns();
- } while (unlikely(read_seqretry(&gtod->lock, seq)));
+ } while (unlikely(read_seqcount_retry(&gtod->seq, seq)));
timespec_add_ns(ts, ns);
return 0;
}
@@ -99,12 +99,12 @@ notrace static noinline int do_monotonic(struct timespec *ts)
{
unsigned long seq, ns, secs;
do {
- seq = read_seqbegin(&gtod->lock);
+ seq = read_seqcount_begin(&gtod->seq);
secs = gtod->wall_time_sec;
ns = gtod->wall_time_nsec + vgetns();
secs += gtod->wall_to_monotonic.tv_sec;
ns += gtod->wall_to_monotonic.tv_nsec;
- } while (unlikely(read_seqretry(&gtod->lock, seq)));
+ } while (unlikely(read_seqcount_retry(&gtod->seq, seq)));
/* wall_time_nsec, vgetns(), and wall_to_monotonic.tv_nsec
* are all guaranteed to be nonnegative.
@@ -123,10 +123,10 @@ notrace static noinline int do_realtime_coarse(struct timespec *ts)
{
unsigned long seq;
do {
- seq = read_seqbegin(&gtod->lock);
+ seq = read_seqcount_begin(&gtod->seq);
ts->tv_sec = gtod->wall_time_coarse.tv_sec;
ts->tv_nsec = gtod->wall_time_coarse.tv_nsec;
- } while (unlikely(read_seqretry(&gtod->lock, seq)));
+ } while (unlikely(read_seqcount_retry(&gtod->seq, seq)));
return 0;
}
@@ -134,12 +134,12 @@ notrace static noinline int do_monotonic_coarse(struct timespec *ts)
{
unsigned long seq, ns, secs;
do {
- seq = read_seqbegin(&gtod->lock);
+ seq = read_seqcount_begin(&gtod->seq);
secs = gtod->wall_time_coarse.tv_sec;
ns = gtod->wall_time_coarse.tv_nsec;
secs += gtod->wall_to_monotonic.tv_sec;
ns += gtod->wall_to_monotonic.tv_nsec;
- } while (unlikely(read_seqretry(&gtod->lock, seq)));
+ } while (unlikely(read_seqcount_retry(&gtod->seq, seq)));
/* wall_time_nsec and wall_to_monotonic.tv_nsec are
* guaranteed to be between 0 and NSEC_PER_SEC.
--
1.7.10

95
debian/patches/features/all/rt/0018-ia64-vsyscall-Use-seqcount-instead-of-seqlock.patch vendored
View File

@ -1,95 +0,0 @@
From 0ad3596103ef2c8ecb6145c8a98ed7e38a9bc6ce Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Feb 2012 18:33:08 +0100
Subject: [PATCH 018/267] ia64: vsyscall: Use seqcount instead of seqlock
The update of the vdso data happens under xtime_lock, so adding a
nested lock is pointless. Just use a seqcount to sync the readers.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
---
arch/ia64/kernel/asm-offsets.c | 4 ++--
arch/ia64/kernel/fsys.S | 2 +-
arch/ia64/kernel/fsyscall_gtod_data.h | 2 +-
arch/ia64/kernel/time.c | 10 +++-------
4 files changed, 7 insertions(+), 11 deletions(-)
diff --git a/arch/ia64/kernel/asm-offsets.c b/arch/ia64/kernel/asm-offsets.c
index af56501..106aeb6 100644
--- a/arch/ia64/kernel/asm-offsets.c
+++ b/arch/ia64/kernel/asm-offsets.c
@@ -269,8 +269,8 @@ void foo(void)
BLANK();
/* used by fsys_gettimeofday in arch/ia64/kernel/fsys.S */
- DEFINE(IA64_GTOD_LOCK_OFFSET,
- offsetof (struct fsyscall_gtod_data_t, lock));
+ DEFINE(IA64_GTOD_SEQ_OFFSET,
+ offsetof (struct fsyscall_gtod_data_t, seq);
DEFINE(IA64_GTOD_WALL_TIME_OFFSET,
offsetof (struct fsyscall_gtod_data_t, wall_time));
DEFINE(IA64_GTOD_MONO_TIME_OFFSET,
diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S
index 331d42b..fa77de7 100644
--- a/arch/ia64/kernel/fsys.S
+++ b/arch/ia64/kernel/fsys.S
@@ -174,7 +174,7 @@ ENTRY(fsys_set_tid_address)
FSYS_RETURN
END(fsys_set_tid_address)
-#if IA64_GTOD_LOCK_OFFSET !=0
+#if IA64_GTOD_SEQ_OFFSET !=0
#error fsys_gettimeofday incompatible with changes to struct fsyscall_gtod_data_t
#endif
#if IA64_ITC_JITTER_OFFSET !=0
diff --git a/arch/ia64/kernel/fsyscall_gtod_data.h b/arch/ia64/kernel/fsyscall_gtod_data.h
index 57d2ee6..146b15b 100644
--- a/arch/ia64/kernel/fsyscall_gtod_data.h
+++ b/arch/ia64/kernel/fsyscall_gtod_data.h
@@ -6,7 +6,7 @@
*/
struct fsyscall_gtod_data_t {
- seqlock_t lock;
+ seqcount_t seq;
struct timespec wall_time;
struct timespec monotonic_time;
cycle_t clk_mask;
diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index 43920de..8e991a0 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -35,9 +35,7 @@
static cycle_t itc_get_cycles(struct clocksource *cs);
-struct fsyscall_gtod_data_t fsyscall_gtod_data = {
- .lock = __SEQLOCK_UNLOCKED(fsyscall_gtod_data.lock),
-};
+struct fsyscall_gtod_data_t fsyscall_gtod_data;
struct itc_jitter_data_t itc_jitter_data;
@@ -460,9 +458,7 @@ void update_vsyscall_tz(void)
void update_vsyscall(struct timespec *wall, struct timespec *wtm,
struct clocksource *c, u32 mult)
{
- unsigned long flags;
-
- write_seqlock_irqsave(&fsyscall_gtod_data.lock, flags);
+ write_seqcount_begin(&fsyscall_gtod_data.seq);
/* copy fsyscall clock data */
fsyscall_gtod_data.clk_mask = c->mask;
@@ -485,6 +481,6 @@ void update_vsyscall(struct timespec *wall, struct timespec *wtm,
fsyscall_gtod_data.monotonic_time.tv_sec++;
}
- write_sequnlock_irqrestore(&fsyscall_gtod_data.lock, flags);
+ write_seqcount_end(&fsyscall_gtod_data.seq);
}
--
1.7.10

91
debian/patches/features/all/rt/0021-vfs-fs_struct-Move-code-out-of-seqcount-write-sectio.patch vendored
View File

@ -1,91 +0,0 @@
From 00f0fcf98b55531834e45b94ad6f8b03737e2ce0 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ZenIV.linux.org.uk>
Date: Thu, 15 Mar 2012 18:39:40 +0000
Subject: [PATCH 021/267] vfs: fs_struct: Move code out of seqcount write
sections
RT cannot disable preemption in the seqcount write sections due to
functions called which take "sleeping" spinlocks.
Move the code out of those sections. It does not need to be there.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
fs/fs_struct.c | 27 +++++++++++++++------------
1 file changed, 15 insertions(+), 12 deletions(-)
diff --git a/fs/fs_struct.c b/fs/fs_struct.c
index 78b519c..f5818c4 100644
--- a/fs/fs_struct.c
+++ b/fs/fs_struct.c
@@ -26,11 +26,11 @@ void set_fs_root(struct fs_struct *fs, struct path *path)
{
struct path old_root;
+ path_get_longterm(path);
spin_lock(&fs->lock);
write_seqcount_begin(&fs->seq);
old_root = fs->root;
fs->root = *path;
- path_get_longterm(path);
write_seqcount_end(&fs->seq);
spin_unlock(&fs->lock);
if (old_root.dentry)
@@ -45,11 +45,11 @@ void set_fs_pwd(struct fs_struct *fs, struct path *path)
{
struct path old_pwd;
+ path_get_longterm(path);
spin_lock(&fs->lock);
write_seqcount_begin(&fs->seq);
old_pwd = fs->pwd;
fs->pwd = *path;
- path_get_longterm(path);
write_seqcount_end(&fs->seq);
spin_unlock(&fs->lock);
@@ -57,6 +57,14 @@ void set_fs_pwd(struct fs_struct *fs, struct path *path)
path_put_longterm(&old_pwd);
}
+static inline int replace_path(struct path *p, const struct path *old, const struct path *new)
+{
+ if (likely(p->dentry != old->dentry || p->mnt != old->mnt))
+ return 0;
+ *p = *new;
+ return 1;
+}
+
void chroot_fs_refs(struct path *old_root, struct path *new_root)
{
struct task_struct *g, *p;
@@ -68,21 +76,16 @@ void chroot_fs_refs(struct path *old_root, struct path *new_root)
task_lock(p);
fs = p->fs;
if (fs) {
+ int hits = 0;
spin_lock(&fs->lock);
write_seqcount_begin(&fs->seq);
- if (fs->root.dentry == old_root->dentry
- && fs->root.mnt == old_root->mnt) {
- path_get_longterm(new_root);
- fs->root = *new_root;
+ hits += replace_path(&fs->root, old_root, new_root);
+ hits += replace_path(&fs->pwd, old_root, new_root);
+ write_seqcount_end(&fs->seq);
+ while (hits--) {
count++;
- }
- if (fs->pwd.dentry == old_root->dentry
- && fs->pwd.mnt == old_root->mnt) {
path_get_longterm(new_root);
- fs->pwd = *new_root;
- count++;
}
- write_seqcount_end(&fs->seq);
spin_unlock(&fs->lock);
}
task_unlock(p);
--
1.7.10

537
debian/patches/features/all/rt/0022-timekeeping-Split-xtime_lock.patch vendored
View File

@ -1,537 +0,0 @@
From 575bc99dec3a6b5121859ec0dc56a400313c500b Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 1 Mar 2012 15:14:06 +0100
Subject: [PATCH 022/267] timekeeping: Split xtime_lock
xtime_lock is going to be split apart in mainline, so we can shorten
the seqcount protected regions and avoid updating seqcount in some
code pathes. This is a straight forward split, so we can avoid the
whole mess with raw seqlocks for RT.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/time/jiffies.c | 4 +-
kernel/time/ntp.c | 24 ++++++++----
kernel/time/tick-common.c | 10 +++--
kernel/time/tick-internal.h | 3 +-
kernel/time/tick-sched.c | 16 +++++---
kernel/time/timekeeping.c | 90 +++++++++++++++++++++++++------------------
6 files changed, 88 insertions(+), 59 deletions(-)
diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c
index a470154..21940eb 100644
--- a/kernel/time/jiffies.c
+++ b/kernel/time/jiffies.c
@@ -74,9 +74,9 @@ u64 get_jiffies_64(void)
u64 ret;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
ret = jiffies_64;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
return ret;
}
EXPORT_SYMBOL(get_jiffies_64);
diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
index 4b85a7a..419cbaa 100644
--- a/kernel/time/ntp.c
+++ b/kernel/time/ntp.c
@@ -358,7 +358,8 @@ static enum hrtimer_restart ntp_leap_second(struct hrtimer *timer)
{
enum hrtimer_restart res = HRTIMER_NORESTART;
- write_seqlock(&xtime_lock);
+ raw_spin_lock(&xtime_lock);
+ write_seqcount_begin(&xtime_seq);
switch (time_state) {
case TIME_OK:
@@ -388,7 +389,8 @@ static enum hrtimer_restart ntp_leap_second(struct hrtimer *timer)
break;
}
- write_sequnlock(&xtime_lock);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock(&xtime_lock);
return res;
}
@@ -663,7 +665,8 @@ int do_adjtimex(struct timex *txc)
getnstimeofday(&ts);
- write_seqlock_irq(&xtime_lock);
+ raw_spin_lock_irq(&xtime_lock);
+ write_seqcount_begin(&xtime_seq);
if (txc->modes & ADJ_ADJTIME) {
long save_adjust = time_adjust;
@@ -705,7 +708,8 @@ int do_adjtimex(struct timex *txc)
/* fill PPS status fields */
pps_fill_timex(txc);
- write_sequnlock_irq(&xtime_lock);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irq(&xtime_lock);
txc->time.tv_sec = ts.tv_sec;
txc->time.tv_usec = ts.tv_nsec;
@@ -903,7 +907,8 @@ void hardpps(const struct timespec *phase_ts, const struct timespec *raw_ts)
pts_norm = pps_normalize_ts(*phase_ts);
- write_seqlock_irqsave(&xtime_lock, flags);
+ raw_spin_lock_irqsave(&xtime_lock, flags);
+ write_seqcount_begin(&xtime_seq);
/* clear the error bits, they will be set again if needed */
time_status &= ~(STA_PPSJITTER | STA_PPSWANDER | STA_PPSERROR);
@@ -916,7 +921,8 @@ void hardpps(const struct timespec *phase_ts, const struct timespec *raw_ts)
* just start the frequency interval */
if (unlikely(pps_fbase.tv_sec == 0)) {
pps_fbase = *raw_ts;
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
return;
}
@@ -931,7 +937,8 @@ void hardpps(const struct timespec *phase_ts, const struct timespec *raw_ts)
time_status |= STA_PPSJITTER;
/* restart the frequency calibration interval */
pps_fbase = *raw_ts;
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
pr_err("hardpps: PPSJITTER: bad pulse\n");
return;
}
@@ -948,7 +955,8 @@ void hardpps(const struct timespec *phase_ts, const struct timespec *raw_ts)
hardpps_update_phase(pts_norm.nsec);
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
}
EXPORT_SYMBOL(hardpps);
diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
index da6c9ec..39de540 100644
--- a/kernel/time/tick-common.c
+++ b/kernel/time/tick-common.c
@@ -63,13 +63,15 @@ int tick_is_oneshot_available(void)
static void tick_periodic(int cpu)
{
if (tick_do_timer_cpu == cpu) {
- write_seqlock(&xtime_lock);
+ raw_spin_lock(&xtime_lock);
+ write_seqcount_begin(&xtime_seq);
/* Keep track of the next tick event */
tick_next_period = ktime_add(tick_next_period, tick_period);
do_timer(1);
- write_sequnlock(&xtime_lock);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock(&xtime_lock);
}
update_process_times(user_mode(get_irq_regs()));
@@ -130,9 +132,9 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
ktime_t next;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
next = tick_next_period;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
clockevents_set_mode(dev, CLOCK_EVT_MODE_ONESHOT);
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index 4e265b9..c91100d 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -141,4 +141,5 @@ static inline int tick_device_is_functional(struct clock_event_device *dev)
#endif
extern void do_timer(unsigned long ticks);
-extern seqlock_t xtime_lock;
+extern raw_spinlock_t xtime_lock;
+extern seqcount_t xtime_seq;
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index c923640..d7abd2f 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -56,7 +56,8 @@ static void tick_do_update_jiffies64(ktime_t now)
return;
/* Reevalute with xtime_lock held */
- write_seqlock(&xtime_lock);
+ raw_spin_lock(&xtime_lock);
+ write_seqcount_begin(&xtime_seq);
delta = ktime_sub(now, last_jiffies_update);
if (delta.tv64 >= tick_period.tv64) {
@@ -79,7 +80,8 @@ static void tick_do_update_jiffies64(ktime_t now)
/* Keep the tick_next_period variable up to date */
tick_next_period = ktime_add(last_jiffies_update, tick_period);
}
- write_sequnlock(&xtime_lock);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock(&xtime_lock);
}
/*
@@ -89,12 +91,14 @@ static ktime_t tick_init_jiffy_update(void)
{
ktime_t period;
- write_seqlock(&xtime_lock);
+ raw_spin_lock(&xtime_lock);
+ write_seqcount_begin(&xtime_seq);
/* Did we start the jiffies update yet ? */
if (last_jiffies_update.tv64 == 0)
last_jiffies_update = tick_next_period;
period = last_jiffies_update;
- write_sequnlock(&xtime_lock);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock(&xtime_lock);
return period;
}
@@ -345,11 +349,11 @@ void tick_nohz_stop_sched_tick(int inidle)
ts->idle_calls++;
/* Read jiffies and the time when jiffies were updated last */
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
last_update = last_jiffies_update;
last_jiffies = jiffies;
time_delta = timekeeping_max_deferment();
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
if (rcu_needs_cpu(cpu) || printk_needs_cpu(cpu) ||
arch_needs_cpu(cpu)) {
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 2378413..da9e1f9 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -139,8 +139,8 @@ static inline s64 timekeeping_get_ns_raw(void)
* This read-write spinlock protects us from races in SMP while
* playing with xtime.
*/
-__cacheline_aligned_in_smp DEFINE_SEQLOCK(xtime_lock);
-
+__cacheline_aligned_in_smp DEFINE_RAW_SPINLOCK(xtime_lock);
+seqcount_t xtime_seq;
/*
* The current time
@@ -222,7 +222,7 @@ void getnstimeofday(struct timespec *ts)
WARN_ON(timekeeping_suspended);
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
*ts = xtime;
nsecs = timekeeping_get_ns();
@@ -230,7 +230,7 @@ void getnstimeofday(struct timespec *ts)
/* If arch requires, add in gettimeoffset() */
nsecs += arch_gettimeoffset();
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
timespec_add_ns(ts, nsecs);
}
@@ -245,14 +245,14 @@ ktime_t ktime_get(void)
WARN_ON(timekeeping_suspended);
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
secs = xtime.tv_sec + wall_to_monotonic.tv_sec;
nsecs = xtime.tv_nsec + wall_to_monotonic.tv_nsec;
nsecs += timekeeping_get_ns();
/* If arch requires, add in gettimeoffset() */
nsecs += arch_gettimeoffset();
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
/*
* Use ktime_set/ktime_add_ns to create a proper ktime on
* 32-bit architectures without CONFIG_KTIME_SCALAR.
@@ -278,14 +278,14 @@ void ktime_get_ts(struct timespec *ts)
WARN_ON(timekeeping_suspended);
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
*ts = xtime;
tomono = wall_to_monotonic;
nsecs = timekeeping_get_ns();
/* If arch requires, add in gettimeoffset() */
nsecs += arch_gettimeoffset();
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec,
ts->tv_nsec + tomono.tv_nsec + nsecs);
@@ -313,7 +313,7 @@ void getnstime_raw_and_real(struct timespec *ts_raw, struct timespec *ts_real)
do {
u32 arch_offset;
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
*ts_raw = raw_time;
*ts_real = xtime;
@@ -326,7 +326,7 @@ void getnstime_raw_and_real(struct timespec *ts_raw, struct timespec *ts_real)
nsecs_raw += arch_offset;
nsecs_real += arch_offset;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
timespec_add_ns(ts_raw, nsecs_raw);
timespec_add_ns(ts_real, nsecs_real);
@@ -365,7 +365,8 @@ int do_settimeofday(const struct timespec *tv)
if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
return -EINVAL;
- write_seqlock_irqsave(&xtime_lock, flags);
+ raw_spin_lock_irqsave(&xtime_lock, flags);
+ write_seqcount_begin(&xtime_seq);
timekeeping_forward_now();
@@ -381,7 +382,8 @@ int do_settimeofday(const struct timespec *tv)
update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock,
timekeeper.mult);
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
/* signal hrtimers about time change */
clock_was_set();
@@ -405,7 +407,8 @@ int timekeeping_inject_offset(struct timespec *ts)
if ((unsigned long)ts->tv_nsec >= NSEC_PER_SEC)
return -EINVAL;
- write_seqlock_irqsave(&xtime_lock, flags);
+ raw_spin_lock_irqsave(&xtime_lock, flags);
+ write_seqcount_begin(&xtime_seq);
timekeeping_forward_now();
@@ -418,7 +421,8 @@ int timekeeping_inject_offset(struct timespec *ts)
update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock,
timekeeper.mult);
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
/* signal hrtimers about time change */
clock_was_set();
@@ -490,11 +494,11 @@ void getrawmonotonic(struct timespec *ts)
s64 nsecs;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
nsecs = timekeeping_get_ns_raw();
*ts = raw_time;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
timespec_add_ns(ts, nsecs);
}
@@ -510,11 +514,11 @@ int timekeeping_valid_for_hres(void)
int ret;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
ret = timekeeper.clock->flags & CLOCK_SOURCE_VALID_FOR_HRES;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
return ret;
}
@@ -572,7 +576,8 @@ void __init timekeeping_init(void)
read_persistent_clock(&now);
read_boot_clock(&boot);
- write_seqlock_irqsave(&xtime_lock, flags);
+ raw_spin_lock_irqsave(&xtime_lock, flags);
+ write_seqcount_begin(&xtime_seq);
ntp_init();
@@ -593,7 +598,8 @@ void __init timekeeping_init(void)
-boot.tv_sec, -boot.tv_nsec);
total_sleep_time.tv_sec = 0;
total_sleep_time.tv_nsec = 0;
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
}
/* time in seconds when suspend began */
@@ -640,7 +646,8 @@ void timekeeping_inject_sleeptime(struct timespec *delta)
if (!(ts.tv_sec == 0 && ts.tv_nsec == 0))
return;
- write_seqlock_irqsave(&xtime_lock, flags);
+ raw_spin_lock_irqsave(&xtime_lock, flags);
+ write_seqcount_begin(&xtime_seq);
timekeeping_forward_now();
__timekeeping_inject_sleeptime(delta);
@@ -650,7 +657,8 @@ void timekeeping_inject_sleeptime(struct timespec *delta)
update_vsyscall(&xtime, &wall_to_monotonic, timekeeper.clock,
timekeeper.mult);
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
/* signal hrtimers about time change */
clock_was_set();
@@ -673,7 +681,8 @@ static void timekeeping_resume(void)
clocksource_resume();
- write_seqlock_irqsave(&xtime_lock, flags);
+ raw_spin_lock_irqsave(&xtime_lock, flags);
+ write_seqcount_begin(&xtime_seq);
if (timespec_compare(&ts, &timekeeping_suspend_time) > 0) {
ts = timespec_sub(ts, timekeeping_suspend_time);
@@ -683,7 +692,8 @@ static void timekeeping_resume(void)
timekeeper.clock->cycle_last = timekeeper.clock->read(timekeeper.clock);
timekeeper.ntp_error = 0;
timekeeping_suspended = 0;
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
touch_softlockup_watchdog();
@@ -701,7 +711,8 @@ static int timekeeping_suspend(void)
read_persistent_clock(&timekeeping_suspend_time);
- write_seqlock_irqsave(&xtime_lock, flags);
+ raw_spin_lock_irqsave(&xtime_lock, flags);
+ write_seqcount_begin(&xtime_seq);
timekeeping_forward_now();
timekeeping_suspended = 1;
@@ -724,7 +735,8 @@ static int timekeeping_suspend(void)
timekeeping_suspend_time =
timespec_add(timekeeping_suspend_time, delta_delta);
}
- write_sequnlock_irqrestore(&xtime_lock, flags);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock_irqrestore(&xtime_lock, flags);
clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL);
clocksource_suspend();
@@ -1101,13 +1113,13 @@ void get_monotonic_boottime(struct timespec *ts)
WARN_ON(timekeeping_suspended);
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
*ts = xtime;
tomono = wall_to_monotonic;
sleep = total_sleep_time;
nsecs = timekeeping_get_ns();
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
set_normalized_timespec(ts, ts->tv_sec + tomono.tv_sec + sleep.tv_sec,
ts->tv_nsec + tomono.tv_nsec + sleep.tv_nsec + nsecs);
@@ -1158,10 +1170,10 @@ struct timespec current_kernel_time(void)
unsigned long seq;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
now = xtime;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
return now;
}
@@ -1173,11 +1185,11 @@ struct timespec get_monotonic_coarse(void)
unsigned long seq;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
now = xtime;
mono = wall_to_monotonic;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
set_normalized_timespec(&now, now.tv_sec + mono.tv_sec,
now.tv_nsec + mono.tv_nsec);
@@ -1209,11 +1221,11 @@ void get_xtime_and_monotonic_and_sleep_offset(struct timespec *xtim,
unsigned long seq;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
*xtim = xtime;
*wtom = wall_to_monotonic;
*sleep = total_sleep_time;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
}
/**
@@ -1225,9 +1237,9 @@ ktime_t ktime_get_monotonic_offset(void)
struct timespec wtom;
do {
- seq = read_seqbegin(&xtime_lock);
+ seq = read_seqcount_begin(&xtime_seq);
wtom = wall_to_monotonic;
- } while (read_seqretry(&xtime_lock, seq));
+ } while (read_seqcount_retry(&xtime_seq, seq));
return timespec_to_ktime(wtom);
}
@@ -1239,7 +1251,9 @@ ktime_t ktime_get_monotonic_offset(void)
*/
void xtime_update(unsigned long ticks)
{
- write_seqlock(&xtime_lock);
+ raw_spin_lock(&xtime_lock);
+ write_seqcount_begin(&xtime_seq);
do_timer(ticks);
- write_sequnlock(&xtime_lock);
+ write_seqcount_end(&xtime_seq);
+ raw_spin_unlock(&xtime_lock);
}
--
1.7.10

125
debian/patches/features/all/rt/0024-mm-memcg-shorten-preempt-disabled-section-around-eve.patch vendored
View File

@ -1,125 +0,0 @@
From 2d38d140493b8b52deebbe11692a71e4513b7605 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@cmpxchg.org>
Date: Thu, 17 Nov 2011 07:49:25 +0100
Subject: [PATCH 024/267] mm: memcg: shorten preempt-disabled section around
event checks
Only the ratelimit checks themselves have to run with preemption
disabled, the resulting actions - checking for usage thresholds,
updating the soft limit tree - can and should run with preemption
enabled.
Signed-off-by: Johannes Weiner <jweiner@redhat.com>
Tested-by: Luis Henriques <henrix@camandro.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
mm/memcontrol.c | 73 ++++++++++++++++++++++++++-----------------------------
1 file changed, 35 insertions(+), 38 deletions(-)
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 778554f..abf5497 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -683,37 +683,32 @@ static unsigned long mem_cgroup_nr_lru_pages(struct mem_cgroup *memcg,
return total;
}
-static bool __memcg_event_check(struct mem_cgroup *memcg, int target)
+static bool mem_cgroup_event_ratelimit(struct mem_cgroup *memcg,
+ enum mem_cgroup_events_target target)
{
unsigned long val, next;
val = __this_cpu_read(memcg->stat->events[MEM_CGROUP_EVENTS_COUNT]);
next = __this_cpu_read(memcg->stat->targets[target]);
/* from time_after() in jiffies.h */
- return ((long)next - (long)val < 0);
-}
-
-static void __mem_cgroup_target_update(struct mem_cgroup *memcg, int target)
-{
- unsigned long val, next;
-
- val = __this_cpu_read(memcg->stat->events[MEM_CGROUP_EVENTS_COUNT]);
-
- switch (target) {
- case MEM_CGROUP_TARGET_THRESH:
- next = val + THRESHOLDS_EVENTS_TARGET;
- break;
- case MEM_CGROUP_TARGET_SOFTLIMIT:
- next = val + SOFTLIMIT_EVENTS_TARGET;
- break;
- case MEM_CGROUP_TARGET_NUMAINFO:
- next = val + NUMAINFO_EVENTS_TARGET;
- break;
- default:
- return;
+ if ((long)next - (long)val < 0) {
+ switch (target) {
+ case MEM_CGROUP_TARGET_THRESH:
+ next = val + THRESHOLDS_EVENTS_TARGET;
+ break;
+ case MEM_CGROUP_TARGET_SOFTLIMIT:
+ next = val + SOFTLIMIT_EVENTS_TARGET;
+ break;
+ case MEM_CGROUP_TARGET_NUMAINFO:
+ next = val + NUMAINFO_EVENTS_TARGET;
+ break;
+ default:
+ break;
+ }
+ __this_cpu_write(memcg->stat->targets[target], next);
+ return true;
}
-
- __this_cpu_write(memcg->stat->targets[target], next);
+ return false;
}
/*
@@ -724,25 +719,27 @@ static void memcg_check_events(struct mem_cgroup *memcg, struct page *page)
{
preempt_disable();
/* threshold event is triggered in finer grain than soft limit */
- if (unlikely(__memcg_event_check(memcg, MEM_CGROUP_TARGET_THRESH))) {
+ if (unlikely(mem_cgroup_event_ratelimit(memcg,
+ MEM_CGROUP_TARGET_THRESH))) {
+ bool do_softlimit, do_numainfo;
+
+ do_softlimit = mem_cgroup_event_ratelimit(memcg,
+ MEM_CGROUP_TARGET_SOFTLIMIT);
+#if MAX_NUMNODES > 1
+ do_numainfo = mem_cgroup_event_ratelimit(memcg,
+ MEM_CGROUP_TARGET_NUMAINFO);
+#endif
+ preempt_enable();
+
mem_cgroup_threshold(memcg);
- __mem_cgroup_target_update(memcg, MEM_CGROUP_TARGET_THRESH);
- if (unlikely(__memcg_event_check(memcg,
- MEM_CGROUP_TARGET_SOFTLIMIT))) {
+ if (unlikely(do_softlimit))
mem_cgroup_update_tree(memcg, page);
- __mem_cgroup_target_update(memcg,
- MEM_CGROUP_TARGET_SOFTLIMIT);
- }
#if MAX_NUMNODES > 1
- if (unlikely(__memcg_event_check(memcg,
- MEM_CGROUP_TARGET_NUMAINFO))) {
+ if (unlikely(do_numainfo))
atomic_inc(&memcg->numainfo_events);
- __mem_cgroup_target_update(memcg,
- MEM_CGROUP_TARGET_NUMAINFO);
- }
#endif
- }
- preempt_enable();
+ } else
+ preempt_enable();
}
static struct mem_cgroup *mem_cgroup_from_cont(struct cgroup *cont)
--
1.7.10

55
debian/patches/features/all/rt/0030-sched-Create-schedule_preempt_disabled.patch vendored
View File

@ -1,55 +0,0 @@
From 64b759067a99f8482feb0c2f75f6ce0fde842166 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 21 Mar 2011 12:09:35 +0100
Subject: [PATCH 030/267] sched: Create schedule_preempt_disabled()
Get rid of the ever repeating:
preempt_enable_no_resched();
schedule();
preempt_disable();
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/sched.h | 1 +
kernel/sched.c | 12 ++++++++++++
2 files changed, 13 insertions(+)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 1c4f3e9..9897fe6 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -359,6 +359,7 @@ extern signed long schedule_timeout_interruptible(signed long timeout);
extern signed long schedule_timeout_killable(signed long timeout);
extern signed long schedule_timeout_uninterruptible(signed long timeout);
asmlinkage void schedule(void);
+extern void schedule_preempt_disabled(void);
extern int mutex_spin_on_owner(struct mutex *lock, struct task_struct *owner);
struct nsproxy;
diff --git a/kernel/sched.c b/kernel/sched.c
index 6b32395..465fd11 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4456,6 +4456,18 @@ asmlinkage void __sched schedule(void)
}
EXPORT_SYMBOL(schedule);
+/**
+ * schedule_preempt_disabled - called with preemption disabled
+ *
+ * Returns with preemption disabled. Note: preempt_count must be 1
+ */
+void __sched schedule_preempt_disabled(void)
+{
+ preempt_enable_no_resched();
+ schedule();
+ preempt_disable();
+}
+
#ifdef CONFIG_MUTEX_SPIN_ON_OWNER
static inline bool owner_running(struct mutex *lock, struct task_struct *owner)
--
1.7.10

500
debian/patches/features/all/rt/0031-sched-Use-schedule_preempt_disabled.patch vendored
View File

@ -1,500 +0,0 @@
From 662df03fb70d4e277547eaa2adfd7e0d520bb35f Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 21 Mar 2011 12:33:18 +0100
Subject: [PATCH 031/267] sched: Use schedule_preempt_disabled()
Coccinelle based conversion.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/kernel/process.c | 4 +---
arch/avr32/kernel/process.c | 4 +---
arch/blackfin/kernel/process.c | 4 +---
arch/cris/kernel/process.c | 4 +---
arch/frv/kernel/process.c | 4 +---
arch/h8300/kernel/process.c | 4 +---
arch/ia64/kernel/process.c | 4 +---
arch/m32r/kernel/process.c | 4 +---
arch/m68k/kernel/process_mm.c | 4 +---
arch/m68k/kernel/process_no.c | 4 +---
arch/microblaze/kernel/process.c | 4 +---
arch/mips/kernel/process.c | 4 +---
arch/mn10300/kernel/process.c | 4 +---
arch/parisc/kernel/process.c | 4 +---
arch/powerpc/kernel/idle.c | 8 ++++----
arch/powerpc/platforms/iseries/setup.c | 8 ++------
arch/s390/kernel/process.c | 4 +---
arch/score/kernel/process.c | 4 +---
arch/sh/kernel/idle.c | 4 +---
arch/sparc/kernel/process_32.c | 8 ++------
arch/sparc/kernel/process_64.c | 10 ++++------
arch/tile/kernel/process.c | 4 +---
arch/x86/kernel/process_32.c | 4 +---
arch/x86/kernel/process_64.c | 4 +---
arch/xtensa/kernel/process.c | 4 +---
init/main.c | 5 +----
kernel/mutex.c | 4 +---
kernel/softirq.c | 4 +---
28 files changed, 36 insertions(+), 95 deletions(-)
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 3d0c6fb..54833ff 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -214,9 +214,7 @@ void cpu_idle(void)
}
leds_event(led_idle_end);
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
index ef5a2a0..c8724c9 100644
--- a/arch/avr32/kernel/process.c
+++ b/arch/avr32/kernel/process.c
@@ -38,9 +38,7 @@ void cpu_idle(void)
while (!need_resched())
cpu_idle_sleep();
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
index 6a80a9e..11acc10 100644
--- a/arch/blackfin/kernel/process.c
+++ b/arch/blackfin/kernel/process.c
@@ -92,9 +92,7 @@ void cpu_idle(void)
while (!need_resched())
idle();
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/cris/kernel/process.c b/arch/cris/kernel/process.c
index aa585e4..d8f50ff 100644
--- a/arch/cris/kernel/process.c
+++ b/arch/cris/kernel/process.c
@@ -115,9 +115,7 @@ void cpu_idle (void)
idle = default_idle;
idle();
}
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/frv/kernel/process.c b/arch/frv/kernel/process.c
index 3901df1..29cc497 100644
--- a/arch/frv/kernel/process.c
+++ b/arch/frv/kernel/process.c
@@ -92,9 +92,7 @@ void cpu_idle(void)
idle();
}
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c
index 933bd38..1a173b3 100644
--- a/arch/h8300/kernel/process.c
+++ b/arch/h8300/kernel/process.c
@@ -81,9 +81,7 @@ void cpu_idle(void)
while (1) {
while (!need_resched())
idle();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c
index 6d33c5c..9dc52b6 100644
--- a/arch/ia64/kernel/process.c
+++ b/arch/ia64/kernel/process.c
@@ -330,9 +330,7 @@ cpu_idle (void)
normal_xtp();
#endif
}
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
check_pgt_cache();
if (cpu_is_offline(cpu))
play_dead();
diff --git a/arch/m32r/kernel/process.c b/arch/m32r/kernel/process.c
index 422bea9..3a4a32b 100644
--- a/arch/m32r/kernel/process.c
+++ b/arch/m32r/kernel/process.c
@@ -90,9 +90,7 @@ void cpu_idle (void)
idle();
}
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/m68k/kernel/process_mm.c b/arch/m68k/kernel/process_mm.c
index aa4ffb8..c413aa0 100644
--- a/arch/m68k/kernel/process_mm.c
+++ b/arch/m68k/kernel/process_mm.c
@@ -94,9 +94,7 @@ void cpu_idle(void)
while (1) {
while (!need_resched())
idle();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/m68k/kernel/process_no.c b/arch/m68k/kernel/process_no.c
index 5e1078c..f7fe6c3 100644
--- a/arch/m68k/kernel/process_no.c
+++ b/arch/m68k/kernel/process_no.c
@@ -73,9 +73,7 @@ void cpu_idle(void)
/* endless idle loop with no priority at all */
while (1) {
idle();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/microblaze/kernel/process.c b/arch/microblaze/kernel/process.c
index 95cc295..d3b2b42 100644
--- a/arch/microblaze/kernel/process.c
+++ b/arch/microblaze/kernel/process.c
@@ -108,9 +108,7 @@ void cpu_idle(void)
idle();
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
check_pgt_cache();
}
}
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index c47f96e..4dbf66d 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -78,9 +78,7 @@ void __noreturn cpu_idle(void)
play_dead();
#endif
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/mn10300/kernel/process.c b/arch/mn10300/kernel/process.c
index 28eec31..cac401d 100644
--- a/arch/mn10300/kernel/process.c
+++ b/arch/mn10300/kernel/process.c
@@ -123,9 +123,7 @@ void cpu_idle(void)
idle();
}
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c
index 4b4b918..f6eb367 100644
--- a/arch/parisc/kernel/process.c
+++ b/arch/parisc/kernel/process.c
@@ -71,9 +71,7 @@ void cpu_idle(void)
while (1) {
while (!need_resched())
barrier();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
check_pgt_cache();
}
}
diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c
index 39a2baa..f46dae5 100644
--- a/arch/powerpc/kernel/idle.c
+++ b/arch/powerpc/kernel/idle.c
@@ -94,11 +94,11 @@ void cpu_idle(void)
HMT_medium();
ppc64_runlatch_on();
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- if (cpu_should_die())
+ if (cpu_should_die()) {
+ preempt_enable_no_resched();
cpu_die();
- schedule();
- preempt_disable();
+ }
+ schedule_preempt_disabled();
}
}
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index ea0acbd..e0c5b49 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -582,9 +582,7 @@ static void iseries_shared_idle(void)
if (hvlpevent_is_pending())
process_iSeries_events();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
@@ -611,9 +609,7 @@ static void iseries_dedicated_idle(void)
ppc64_runlatch_on();
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 53088e2..fa093f7 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -94,9 +94,7 @@ void cpu_idle(void)
while (!need_resched())
default_idle();
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/score/kernel/process.c b/arch/score/kernel/process.c
index 25d0803..2707023 100644
--- a/arch/score/kernel/process.c
+++ b/arch/score/kernel/process.c
@@ -53,9 +53,7 @@ void __noreturn cpu_idle(void)
while (!need_resched())
barrier();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/sh/kernel/idle.c b/arch/sh/kernel/idle.c
index db4ecd7..b7c18f0 100644
--- a/arch/sh/kernel/idle.c
+++ b/arch/sh/kernel/idle.c
@@ -112,9 +112,7 @@ void cpu_idle(void)
}
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/sparc/kernel/process_32.c b/arch/sparc/kernel/process_32.c
index f793742..935fdbc 100644
--- a/arch/sparc/kernel/process_32.c
+++ b/arch/sparc/kernel/process_32.c
@@ -113,9 +113,7 @@ void cpu_idle(void)
while (!need_resched())
cpu_relax();
}
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
check_pgt_cache();
}
}
@@ -138,9 +136,7 @@ void cpu_idle(void)
while (!need_resched())
cpu_relax();
}
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
check_pgt_cache();
}
}
diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
index 3739a06..8ba0dbe 100644
--- a/arch/sparc/kernel/process_64.c
+++ b/arch/sparc/kernel/process_64.c
@@ -102,15 +102,13 @@ void cpu_idle(void)
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
-
#ifdef CONFIG_HOTPLUG_CPU
- if (cpu_is_offline(cpu))
+ if (cpu_is_offline(cpu)) {
+ preempt_enable_no_resched();
cpu_play_dead();
+ }
#endif
-
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/tile/kernel/process.c b/arch/tile/kernel/process.c
index 9c45d8b..5a8b631 100644
--- a/arch/tile/kernel/process.c
+++ b/arch/tile/kernel/process.c
@@ -106,9 +106,7 @@ void cpu_idle(void)
current_thread_info()->status |= TS_POLLING;
}
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index 8598296..ada175e3 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -117,9 +117,7 @@ void cpu_idle(void)
start_critical_timings();
}
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index 6a364a6..08840ab 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -150,9 +150,7 @@ void cpu_idle(void)
}
tick_nohz_restart_sched_tick();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/arch/xtensa/kernel/process.c b/arch/xtensa/kernel/process.c
index 47041e7..2c90047 100644
--- a/arch/xtensa/kernel/process.c
+++ b/arch/xtensa/kernel/process.c
@@ -113,9 +113,7 @@ void cpu_idle(void)
while (1) {
while (!need_resched())
platform_idle();
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
}
diff --git a/init/main.c b/init/main.c
index 217ed23..0e274f9 100644
--- a/init/main.c
+++ b/init/main.c
@@ -378,11 +378,8 @@ static noinline void __init_refok rest_init(void)
* at least once to get things moving:
*/
init_idle_bootup_task(current);
- preempt_enable_no_resched();
- schedule();
-
+ schedule_preempt_disabled();
/* Call into cpu_idle with preempt disabled */
- preempt_disable();
cpu_idle();
}
diff --git a/kernel/mutex.c b/kernel/mutex.c
index 89096dd..a307cc9 100644
--- a/kernel/mutex.c
+++ b/kernel/mutex.c
@@ -240,9 +240,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
/* didn't get the lock, go to sleep: */
spin_unlock_mutex(&lock->wait_lock, flags);
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
spin_lock_mutex(&lock->wait_lock, flags);
}
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 2c71d91..c9da2c8 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -744,9 +744,7 @@ static int run_ksoftirqd(void * __bind_cpu)
while (!kthread_should_stop()) {
preempt_disable();
if (!local_softirq_pending()) {
- preempt_enable_no_resched();
- schedule();
- preempt_disable();
+ schedule_preempt_disabled();
}
__set_current_state(TASK_RUNNING);
--
1.7.10

71
debian/patches/features/all/rt/0038-ARM-AT91-PIT-Remove-irq-handler-when-clock-event-is-.patch vendored
View File

@ -1,71 +0,0 @@
From 31e34bf84003d53073f3fa91d53cd6ac31bfbe58 Mon Sep 17 00:00:00 2001
From: Benedikt Spranger <b.spranger@linutronix.de>
Date: Sat, 6 Mar 2010 17:47:10 +0100
Subject: [PATCH 038/267] ARM: AT91: PIT: Remove irq handler when clock event
is unused
Setup and remove the interrupt handler in clock event mode selection.
This avoids calling the (shared) interrupt handler when the device is
not used.
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/mach-at91/at91rm9200_time.c | 2 ++
arch/arm/mach-at91/at91sam926x_time.c | 6 +++++-
2 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/arch/arm/mach-at91/at91rm9200_time.c b/arch/arm/mach-at91/at91rm9200_time.c
index 1dd69c8..0666570 100644
--- a/arch/arm/mach-at91/at91rm9200_time.c
+++ b/arch/arm/mach-at91/at91rm9200_time.c
@@ -114,6 +114,7 @@ clkevt32k_mode(enum clock_event_mode mode, struct clock_event_device *dev)
last_crtr = read_CRTR();
switch (mode) {
case CLOCK_EVT_MODE_PERIODIC:
+ setup_irq(AT91_ID_SYS, &at91rm9200_timer_irq);
/* PIT for periodic irqs; fixed rate of 1/HZ */
irqmask = AT91_ST_PITS;
at91_sys_write(AT91_ST_PIMR, LATCH);
@@ -127,6 +128,7 @@ clkevt32k_mode(enum clock_event_mode mode, struct clock_event_device *dev)
break;
case CLOCK_EVT_MODE_SHUTDOWN:
case CLOCK_EVT_MODE_UNUSED:
+ remove_irq(AT91_ID_SYS, &at91rm9200_timer_irq);
case CLOCK_EVT_MODE_RESUME:
irqmask = 0;
break;
diff --git a/arch/arm/mach-at91/at91sam926x_time.c b/arch/arm/mach-at91/at91sam926x_time.c
index 4ba8549..97d1e14 100644
--- a/arch/arm/mach-at91/at91sam926x_time.c
+++ b/arch/arm/mach-at91/at91sam926x_time.c
@@ -54,7 +54,7 @@ static struct clocksource pit_clk = {
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
};
-
+static struct irqaction at91sam926x_pit_irq;
/*
* Clockevent device: interrupts every 1/HZ (== pit_cycles * MCK/16)
*/
@@ -63,6 +63,9 @@ pit_clkevt_mode(enum clock_event_mode mode, struct clock_event_device *dev)
{
switch (mode) {
case CLOCK_EVT_MODE_PERIODIC:
+ /* Set up irq handler */
+ setup_irq(AT91_ID_SYS, &at91sam926x_pit_irq);
+
/* update clocksource counter */
pit_cnt += pit_cycle * PIT_PICNT(at91_sys_read(AT91_PIT_PIVR));
at91_sys_write(AT91_PIT_MR, (pit_cycle - 1) | AT91_PIT_PITEN
@@ -75,6 +78,7 @@ pit_clkevt_mode(enum clock_event_mode mode, struct clock_event_device *dev)
case CLOCK_EVT_MODE_UNUSED:
/* disable irq, leaving the clocksource active */
at91_sys_write(AT91_PIT_MR, (pit_cycle - 1) | AT91_PIT_PITEN);
+ remove_irq(AT91_ID_SYS, &at91sam926x_pit_irq);
break;
case CLOCK_EVT_MODE_RESUME:
break;
--
1.7.10

124
debian/patches/features/all/rt/0044-preempt-mark-legitimated-no-resched-sites.patch.patch vendored
View File

@ -1,124 +0,0 @@
From a5f63f3c98dcea7b51fda0df14ecc187fb329eee Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 21 Mar 2011 13:32:17 +0100
Subject: [PATCH 044/267] preempt-mark-legitimated-no-resched-sites.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/powerpc/kernel/idle.c | 2 +-
arch/sparc/kernel/process_64.c | 2 +-
include/linux/preempt.h | 5 ++++-
kernel/sched.c | 6 +++---
kernel/softirq.c | 4 ++--
5 files changed, 11 insertions(+), 8 deletions(-)
diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c
index f46dae5..5d70d10 100644
--- a/arch/powerpc/kernel/idle.c
+++ b/arch/powerpc/kernel/idle.c
@@ -95,7 +95,7 @@ void cpu_idle(void)
ppc64_runlatch_on();
tick_nohz_restart_sched_tick();
if (cpu_should_die()) {
- preempt_enable_no_resched();
+ __preempt_enable_no_resched();
cpu_die();
}
schedule_preempt_disabled();
diff --git a/arch/sparc/kernel/process_64.c b/arch/sparc/kernel/process_64.c
index 8ba0dbe..86fe09a 100644
--- a/arch/sparc/kernel/process_64.c
+++ b/arch/sparc/kernel/process_64.c
@@ -104,7 +104,7 @@ void cpu_idle(void)
#ifdef CONFIG_HOTPLUG_CPU
if (cpu_is_offline(cpu)) {
- preempt_enable_no_resched();
+ __preempt_enable_no_resched();
cpu_play_dead();
}
#endif
diff --git a/include/linux/preempt.h b/include/linux/preempt.h
index 58969b2..227b0f5 100644
--- a/include/linux/preempt.h
+++ b/include/linux/preempt.h
@@ -48,12 +48,14 @@ do { \
barrier(); \
} while (0)
-#define preempt_enable_no_resched() \
+#define __preempt_enable_no_resched() \
do { \
barrier(); \
dec_preempt_count(); \
} while (0)
+#define preempt_enable_no_resched() __preempt_enable_no_resched()
+
#define preempt_enable() \
do { \
preempt_enable_no_resched(); \
@@ -92,6 +94,7 @@ do { \
#else /* !CONFIG_PREEMPT_COUNT */
#define preempt_disable() do { } while (0)
+#define __preempt_enable_no_resched() do { } while (0)
#define preempt_enable_no_resched() do { } while (0)
#define preempt_enable() do { } while (0)
diff --git a/kernel/sched.c b/kernel/sched.c
index 465fd11..e35f01c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4415,7 +4415,7 @@ need_resched:
post_schedule(rq);
- preempt_enable_no_resched();
+ __preempt_enable_no_resched();
if (need_resched())
goto need_resched;
}
@@ -4463,7 +4463,7 @@ EXPORT_SYMBOL(schedule);
*/
void __sched schedule_preempt_disabled(void)
{
- preempt_enable_no_resched();
+ __preempt_enable_no_resched();
schedule();
preempt_disable();
}
@@ -5705,7 +5705,7 @@ SYSCALL_DEFINE0(sched_yield)
__release(rq->lock);
spin_release(&rq->lock.dep_map, 1, _THIS_IP_);
do_raw_spin_unlock(&rq->lock);
- preempt_enable_no_resched();
+ __preempt_enable_no_resched();
schedule();
diff --git a/kernel/softirq.c b/kernel/softirq.c
index c9da2c8..a8becbf 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -353,7 +353,7 @@ void irq_exit(void)
if (idle_cpu(smp_processor_id()) && !in_interrupt() && !need_resched())
tick_nohz_stop_sched_tick(0);
#endif
- preempt_enable_no_resched();
+ __preempt_enable_no_resched();
}
/*
@@ -759,7 +759,7 @@ static int run_ksoftirqd(void * __bind_cpu)
if (local_softirq_pending())
__do_softirq();
local_irq_enable();
- preempt_enable_no_resched();
+ __preempt_enable_no_resched();
cond_resched();
preempt_disable();
rcu_note_context_switch((long)__bind_cpu);
--
1.7.10

40
debian/patches/features/all/rt/0061-softirq-thread-do-softirq.patch.patch vendored
View File

@ -1,40 +0,0 @@
From 96af58acea4b1f2d36ff57313a95c5cd6e8c0e9c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 28 Jun 2011 15:44:15 +0200
Subject: [PATCH 061/267] softirq-thread-do-softirq.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/interrupt.h | 2 ++
net/core/dev.c | 2 +-
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index a64b00e..21b94de 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -454,6 +454,8 @@ struct softirq_action
asmlinkage void do_softirq(void);
asmlinkage void __do_softirq(void);
+static inline void thread_do_softirq(void) { do_softirq(); }
+
extern void open_softirq(int nr, void (*action)(struct softirq_action *));
extern void softirq_init(void);
static inline void __raise_softirq_irqoff(unsigned int nr)
diff --git a/net/core/dev.c b/net/core/dev.c
index 55cd370..c28987a 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3037,7 +3037,7 @@ int netif_rx_ni(struct sk_buff *skb)
preempt_disable();
err = netif_rx(skb);
if (local_softirq_pending())
- do_softirq();
+ thread_do_softirq();
preempt_enable();
return err;
--
1.7.10

41
debian/patches/features/all/rt/0064-x86-32-fix-signal-crap.patch.patch vendored
View File

@ -1,41 +0,0 @@
From 52c45c824d10baf0dd84cc56dc2fc35536d4fa49 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 18 Jul 2011 15:59:38 +0200
Subject: [PATCH 064/267] x86-32-fix-signal-crap.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/kernel/entry_32.S | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S
index bcda816..426cf51 100644
--- a/arch/x86/kernel/entry_32.S
+++ b/arch/x86/kernel/entry_32.S
@@ -629,7 +629,11 @@ work_notifysig: # deal with pending signals and
jne work_notifysig_v86 # returning to kernel-space or
# vm86-space
xorl %edx, %edx
+ TRACE_IRQS_ON
+ ENABLE_INTERRUPTS(CLBR_NONE)
call do_notify_resume
+ DISABLE_INTERRUPTS(CLBR_ANY)
+ TRACE_IRQS_OFF
jmp resume_userspace_sig
ALIGN
@@ -642,7 +646,11 @@ work_notifysig_v86:
movl %esp, %eax
#endif
xorl %edx, %edx
+ TRACE_IRQS_ON
+ ENABLE_INTERRUPTS(CLBR_NONE)
call do_notify_resume
+ DISABLE_INTERRUPTS(CLBR_ANY)
+ TRACE_IRQS_OFF
jmp resume_userspace_sig
END(work_pending)
--
1.7.10

98
debian/patches/features/all/rt/0066-rcu-Reduce-lock-section.patch vendored
View File

@ -1,98 +0,0 @@
From e48b25d1e88f0066fc74583020ad764cd0d002f5 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 24 Jun 2011 22:23:02 +0200
Subject: [PATCH 066/267] rcu: Reduce lock section
So the waitqueue wakeup is outside the raw locked section.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/rcutree.c | 2 +-
kernel/rcutree.h | 3 ++-
kernel/rcutree_plugin.h | 14 ++++++++------
3 files changed, 11 insertions(+), 8 deletions(-)
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 6b76d81..8ef8675 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -1221,7 +1221,7 @@ static void __rcu_offline_cpu(int cpu, struct rcu_state *rsp)
else
raw_spin_unlock_irqrestore(&rnp->lock, flags);
if (need_report & RCU_OFL_TASKS_EXP_GP)
- rcu_report_exp_rnp(rsp, rnp);
+ rcu_report_exp_rnp(rsp, rnp, true);
rcu_node_kthread_setaffinity(rnp, -1);
}
diff --git a/kernel/rcutree.h b/kernel/rcutree.h
index 849ce9e..dca495d 100644
--- a/kernel/rcutree.h
+++ b/kernel/rcutree.h
@@ -451,7 +451,8 @@ static void rcu_preempt_check_callbacks(int cpu);
static void rcu_preempt_process_callbacks(void);
void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu));
#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_TREE_PREEMPT_RCU)
-static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp);
+static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp,
+ bool wake);
#endif /* #if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_TREE_PREEMPT_RCU) */
static int rcu_preempt_pending(int cpu);
static int rcu_preempt_needs_cpu(int cpu);
diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h
index 4b9b9f8..73cab33 100644
--- a/kernel/rcutree_plugin.h
+++ b/kernel/rcutree_plugin.h
@@ -407,7 +407,7 @@ static noinline void rcu_read_unlock_special(struct task_struct *t)
* then we need to report up the rcu_node hierarchy.
*/
if (!empty_exp && !rcu_preempted_readers_exp(rnp))
- rcu_report_exp_rnp(&rcu_preempt_state, rnp);
+ rcu_report_exp_rnp(&rcu_preempt_state, rnp, true);
} else {
local_irq_restore(flags);
}
@@ -731,7 +731,8 @@ static int sync_rcu_preempt_exp_done(struct rcu_node *rnp)
*
* Caller must hold sync_rcu_preempt_exp_mutex.
*/
-static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp)
+static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp,
+ bool wake)
{
unsigned long flags;
unsigned long mask;
@@ -744,7 +745,8 @@ static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp)
}
if (rnp->parent == NULL) {
raw_spin_unlock_irqrestore(&rnp->lock, flags);
- wake_up(&sync_rcu_preempt_exp_wq);
+ if (wake)
+ wake_up(&sync_rcu_preempt_exp_wq);
break;
}
mask = rnp->grpmask;
@@ -777,7 +779,7 @@ sync_rcu_preempt_exp_init(struct rcu_state *rsp, struct rcu_node *rnp)
must_wait = 1;
}
if (!must_wait)
- rcu_report_exp_rnp(rsp, rnp);
+ rcu_report_exp_rnp(rsp, rnp, false);
}
/*
@@ -1069,9 +1071,9 @@ EXPORT_SYMBOL_GPL(synchronize_rcu_expedited);
* report on tasks preempted in RCU read-side critical sections during
* expedited RCU grace periods.
*/
-static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp)
+static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp,
+ bool wake)
{
- return;
}
#endif /* #ifdef CONFIG_HOTPLUG_CPU */
--
1.7.10

58
debian/patches/features/all/rt/0068-wait-Provide-__wake_up_all_locked.patch vendored
View File

@ -1,58 +0,0 @@
From 90713b9e3f210d7ebb96dade4cc71b7397ce99d4 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 1 Dec 2011 00:04:00 +0100
Subject: [PATCH 068/267] wait: Provide __wake_up_all_locked
For code which protects the waitqueue itself with another lock it
makes no sense to acquire the waitqueue lock for wakeup all. Provide
__wake_up_all_locked.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable-rt@vger.kernel.org
---
include/linux/wait.h | 5 +++--
kernel/sched.c | 4 ++--
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/include/linux/wait.h b/include/linux/wait.h
index 3efc9f3..1e904b8 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -157,7 +157,7 @@ void __wake_up(wait_queue_head_t *q, unsigned int mode, int nr, void *key);
void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key);
void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, int nr,
void *key);
-void __wake_up_locked(wait_queue_head_t *q, unsigned int mode);
+void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr);
void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr);
void __wake_up_bit(wait_queue_head_t *, void *, int);
int __wait_on_bit(wait_queue_head_t *, struct wait_bit_queue *, int (*)(void *), unsigned);
@@ -170,7 +170,8 @@ wait_queue_head_t *bit_waitqueue(void *, int);
#define wake_up(x) __wake_up(x, TASK_NORMAL, 1, NULL)
#define wake_up_nr(x, nr) __wake_up(x, TASK_NORMAL, nr, NULL)
#define wake_up_all(x) __wake_up(x, TASK_NORMAL, 0, NULL)
-#define wake_up_locked(x) __wake_up_locked((x), TASK_NORMAL)
+#define wake_up_locked(x) __wake_up_locked((x), TASK_NORMAL, 1)
+#define wake_up_all_locked(x) __wake_up_locked((x), TASK_NORMAL, 0)
#define wake_up_interruptible(x) __wake_up(x, TASK_INTERRUPTIBLE, 1, NULL)
#define wake_up_interruptible_nr(x, nr) __wake_up(x, TASK_INTERRUPTIBLE, nr, NULL)
diff --git a/kernel/sched.c b/kernel/sched.c
index e35f01c..bb93b0b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4637,9 +4637,9 @@ EXPORT_SYMBOL(__wake_up);
/*
* Same as __wake_up but called with the spinlock in wait_queue_head_t held.
*/
-void __wake_up_locked(wait_queue_head_t *q, unsigned int mode)
+void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr)
{
- __wake_up_common(q, mode, 1, 0, NULL);
+ __wake_up_common(q, mode, nr, 0, NULL);
}
EXPORT_SYMBOL_GPL(__wake_up_locked);
--
1.7.10

32
debian/patches/features/all/rt/0069-pci-Use-__wake_up_all_locked-pci_unblock_user_cfg_ac.patch vendored
View File

@ -1,32 +0,0 @@
From 0285604d1f4283b1bb59ec7efd683d4475c018db Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 1 Dec 2011 00:07:16 +0100
Subject: [PATCH 069/267] pci: Use __wake_up_all_locked
pci_unblock_user_cfg_access()
The waitqueue is protected by the pci_lock, so we can just avoid to
lock the waitqueue lock itself. That prevents the
might_sleep()/scheduling while atomic problem on RT
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable-rt@vger.kernel.org
---
drivers/pci/access.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/pci/access.c b/drivers/pci/access.c
index fdaa42a..1a6cc67 100644
--- a/drivers/pci/access.c
+++ b/drivers/pci/access.c
@@ -441,7 +441,7 @@ void pci_unblock_user_cfg_access(struct pci_dev *dev)
WARN_ON(!dev->block_ucfg_access);
dev->block_ucfg_access = 0;
- wake_up_all(&pci_ucfg_wait);
+ wake_up_all_locked(&pci_ucfg_wait);
raw_spin_unlock_irqrestore(&pci_lock, flags);
}
EXPORT_SYMBOL_GPL(pci_unblock_user_cfg_access);
--
1.7.10

23
debian/patches/features/all/rt/0072-localversion.patch.patch vendored
View File

@ -1,23 +0,0 @@
From dc3a0b29103cadebe8c40fd3ad80b18b7d1af27b Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 8 Jul 2011 20:25:16 +0200
Subject: [PATCH 072/267] localversion.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/n/tip-8vdw4bfcsds27cvox6rpb334@git.kernel.org
---
localversion-rt | 1 +
1 file changed, 1 insertion(+)
create mode 100644 localversion-rt
diff --git a/localversion-rt b/localversion-rt
new file mode 100644
index 0000000..b2111a2
--- /dev/null
+++ b/localversion-rt
@@ -0,0 +1 @@
+-rt24
--
1.7.10

98
debian/patches/features/all/rt/0092-mm-Replace-cgroup_page-bit-spinlock.patch vendored
View File

@ -1,98 +0,0 @@
From d553f4dded23aedcf07184908178b715c78cc492 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 19 Aug 2009 09:56:42 +0200
Subject: [PATCH 092/267] mm: Replace cgroup_page bit spinlock
Bit spinlocks are not working on RT. Replace them.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/page_cgroup.h | 28 ++++++++++++++++++++++++++++
mm/page_cgroup.c | 1 +
2 files changed, 29 insertions(+)
diff --git a/include/linux/page_cgroup.h b/include/linux/page_cgroup.h
index 961ecc7..2927c08 100644
--- a/include/linux/page_cgroup.h
+++ b/include/linux/page_cgroup.h
@@ -30,6 +30,10 @@ enum {
*/
struct page_cgroup {
unsigned long flags;
+#ifdef CONFIG_PREEMPT_RT_BASE
+ spinlock_t pcg_lock;
+ spinlock_t pcm_lock;
+#endif
struct mem_cgroup *mem_cgroup;
struct list_head lru; /* per cgroup LRU list */
};
@@ -96,30 +100,54 @@ static inline void lock_page_cgroup(struct page_cgroup *pc)
* Don't take this lock in IRQ context.
* This lock is for pc->mem_cgroup, USED, CACHE, MIGRATION
*/
+#ifndef CONFIG_PREEMPT_RT_BASE
bit_spin_lock(PCG_LOCK, &pc->flags);
+#else
+ spin_lock(&pc->pcg_lock);
+#endif
}
static inline void unlock_page_cgroup(struct page_cgroup *pc)
{
+#ifndef CONFIG_PREEMPT_RT_BASE
bit_spin_unlock(PCG_LOCK, &pc->flags);
+#else
+ spin_unlock(&pc->pcg_lock);
+#endif
}
static inline void move_lock_page_cgroup(struct page_cgroup *pc,
unsigned long *flags)
{
+#ifndef CONFIG_PREEMPT_RT_BASE
/*
* We know updates to pc->flags of page cache's stats are from both of
* usual context or IRQ context. Disable IRQ to avoid deadlock.
*/
local_irq_save(*flags);
bit_spin_lock(PCG_MOVE_LOCK, &pc->flags);
+#else
+ spin_lock_irqsave(&pc->pcm_lock, *flags);
+#endif
}
static inline void move_unlock_page_cgroup(struct page_cgroup *pc,
unsigned long *flags)
{
+#ifndef CONFIG_PREEMPT_RT_BASE
bit_spin_unlock(PCG_MOVE_LOCK, &pc->flags);
local_irq_restore(*flags);
+#else
+ spin_unlock_irqrestore(&pc->pcm_lock, *flags);
+#endif
+}
+
+static inline void page_cgroup_lock_init(struct page_cgroup *pc)
+{
+#ifdef CONFIG_PREEMPT_RT_BASE
+ spin_lock_init(&pc->pcg_lock);
+ spin_lock_init(&pc->pcm_lock);
+#endif
}
#ifdef CONFIG_SPARSEMEM
diff --git a/mm/page_cgroup.c b/mm/page_cgroup.c
index 2d123f9..2e0d18d 100644
--- a/mm/page_cgroup.c
+++ b/mm/page_cgroup.c
@@ -17,6 +17,7 @@ static void __meminit init_page_cgroup(struct page_cgroup *pc, unsigned long id)
set_page_cgroup_array_id(pc, id);
pc->mem_cgroup = NULL;
INIT_LIST_HEAD(&pc->lru);
+ page_cgroup_lock_init(pc);
}
static unsigned long total_usage;
--
1.7.10

56
debian/patches/features/all/rt/0113-mm-vmstat-fix-the-irq-lock-asymetry.patch.patch vendored
View File

@ -1,56 +0,0 @@
From d8ce6bcfe27b0b0649048b054827a7c6a64b5d0c Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 22 Jun 2011 20:47:08 +0200
Subject: [PATCH 113/267] mm-vmstat-fix-the-irq-lock-asymetry.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
mm/vmscan.c | 18 +++++++++---------
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index cb33d9c..39f9629 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1344,8 +1344,8 @@ static int too_many_isolated(struct zone *zone, int file,
*/
static noinline_for_stack void
putback_lru_pages(struct zone *zone, struct scan_control *sc,
- unsigned long nr_anon, unsigned long nr_file,
- struct list_head *page_list)
+ unsigned long nr_anon, unsigned long nr_file,
+ struct list_head *page_list, unsigned long nr_reclaimed)
{
struct page *page;
struct pagevec pvec;
@@ -1356,7 +1356,12 @@ putback_lru_pages(struct zone *zone, struct scan_control *sc,
/*
* Put back any unfreeable pages.
*/
- spin_lock(&zone->lru_lock);
+ spin_lock_irq(&zone->lru_lock);
+
+ if (current_is_kswapd())
+ __count_vm_events(KSWAPD_STEAL, nr_reclaimed);
+ __count_zone_vm_events(PGSTEAL, zone, nr_reclaimed);
+
while (!list_empty(page_list)) {
int lru;
page = lru_to_page(page_list);
@@ -1539,12 +1544,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct zone *zone,
priority, &nr_dirty, &nr_writeback);
}
- local_irq_disable();
- if (current_is_kswapd())
- __count_vm_events(KSWAPD_STEAL, nr_reclaimed);
- __count_zone_vm_events(PGSTEAL, zone, nr_reclaimed);
-
- putback_lru_pages(zone, sc, nr_anon, nr_file, &page_list);
+ putback_lru_pages(zone, sc, nr_anon, nr_file, &page_list, nr_reclaimed);
/*
* If reclaim is isolating dirty pages under writeback, it implies
--
1.7.10

28
debian/patches/features/all/rt/0119-panic-disable-random-on-rt.patch vendored
View File

@ -1,28 +0,0 @@
From f4c10a6646165be65152353cde34f57b80e14519 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 10 Apr 2012 14:34:04 -0400
Subject: [PATCH 119/267] panic-disable-random-on-rt
---
kernel/panic.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/kernel/panic.c b/kernel/panic.c
index 3458469..50f4de5 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -343,9 +343,11 @@ static u64 oops_id;
static int init_oops_id(void)
{
+#ifndef CONFIG_PREEMPT_RT_FULL
if (!oops_id)
get_random_bytes(&oops_id, sizeof(oops_id));
else
+#endif
oops_id++;
return 0;
--
1.7.10

41
debian/patches/features/all/rt/0135-hrtimer-Add-missing-debug_activate-aid-Was-Re-ANNOUN.patch vendored
View File

@ -1,41 +0,0 @@
From 6d0f369ac20fb888983eff59ce6cb1ec93e815c5 Mon Sep 17 00:00:00 2001
From: Yong Zhang <yong.zhang0@gmail.com>
Date: Thu, 13 Oct 2011 15:52:30 +0800
Subject: [PATCH 135/267] hrtimer: Add missing debug_activate() aid [Was: Re:
[ANNOUNCE] 3.0.6-rt17]
On Fri, Oct 07, 2011 at 10:25:25AM -0700, Fernando Lopez-Lezcano wrote:
> On 10/06/2011 06:15 PM, Thomas Gleixner wrote:
> >Dear RT Folks,
> >
> >I'm pleased to announce the 3.0.6-rt17 release.
>
> Hi and thanks again. So far this one is not hanging which is very
> good news. But I still see the hrtimer_fixup_activate warnings I
> reported for rt16...
Hi Fernando,
I think below patch will smooth your concern?
Thanks,
Yong
---
kernel/hrtimer.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 358442b..d363df8 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -1042,6 +1042,7 @@ int __hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
* remove it again and report a failure. This avoids
* stale base->first entries.
*/
+ debug_deactivate(timer);
__remove_hrtimer(timer, new_base,
timer->state & HRTIMER_STATE_CALLBACK, 0);
}
--
1.7.10

55
debian/patches/features/all/rt/0145-sched-prevent-idle-boost.patch.patch vendored
View File

@ -1,55 +0,0 @@
From 2cf8fbc8ec1ba087782bc59ae4e452bd42d6c9e2 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 6 Jun 2011 20:07:38 +0200
Subject: [PATCH 145/267] sched-prevent-idle-boost.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched.c | 21 +++++++++++++++++++--
1 file changed, 19 insertions(+), 2 deletions(-)
diff --git a/kernel/sched.c b/kernel/sched.c
index f101d97..b813946 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5037,6 +5037,24 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
rq = __task_rq_lock(p);
+ /*
+ * Idle task boosting is a nono in general. There is one
+ * exception, when PREEMPT_RT and NOHZ is active:
+ *
+ * The idle task calls get_next_timer_interrupt() and holds
+ * the timer wheel base->lock on the CPU and another CPU wants
+ * to access the timer (probably to cancel it). We can safely
+ * ignore the boosting request, as the idle CPU runs this code
+ * with interrupts disabled and will complete the lock
+ * protected section without being interrupted. So there is no
+ * real need to boost.
+ */
+ if (unlikely(p == rq->idle)) {
+ WARN_ON(p != rq->curr);
+ WARN_ON(p->pi_blocked_on);
+ goto out_unlock;
+ }
+
trace_sched_pi_setprio(p, prio);
oldprio = p->prio;
prev_class = p->sched_class;
@@ -5060,11 +5078,10 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
enqueue_task(rq, p, oldprio < prio ? ENQUEUE_HEAD : 0);
check_class_changed(rq, p, prev_class, oldprio);
+out_unlock:
__task_rq_unlock(rq);
}
-
#endif
-
void set_user_nice(struct task_struct *p, long nice)
{
int old_prio, delta, on_rq;
--
1.7.10

53
debian/patches/features/all/rt/0147-sched-Break-out-from-load_balancing-on-rq_lock-conte.patch vendored
View File

@ -1,53 +0,0 @@
From 99413c8e97ac85b7f0d29603625302a76aa06e1e Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <peterz@infradead.org>
Date: Tue, 16 Mar 2010 14:31:44 -0700
Subject: [PATCH 147/267] sched: Break out from load_balancing on rq_lock
contention
Also limit NEW_IDLE pull
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched_fair.c | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 8a39fa3..3747e53 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -2899,6 +2899,10 @@ balance_tasks(struct rq *this_rq, int this_cpu, struct rq *busiest,
*/
if (idle == CPU_NEWLY_IDLE)
break;
+
+ if (raw_spin_is_contended(&this_rq->lock) ||
+ raw_spin_is_contended(&busiest->lock))
+ break;
#endif
/*
@@ -3039,6 +3043,20 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
rem_load_move -= moved_load;
if (rem_load_move < 0)
break;
+
+#ifdef CONFIG_PREEMPT
+ /*
+ * NEWIDLE balancing is a source of latency, so preemptible
+ * kernels will stop after the first task is pulled to minimize
+ * the critical section.
+ */
+ if (idle == CPU_NEWLY_IDLE && this_rq->nr_running)
+ break;
+
+ if (raw_spin_is_contended(&this_rq->lock) ||
+ raw_spin_is_contended(&busiest->lock))
+ break;
+#endif
}
rcu_read_unlock();
--
1.7.10

62
debian/patches/features/all/rt/0150-sched-no-work-when-pi-blocked.patch.patch vendored
View File

@ -1,62 +0,0 @@
From dd97a1f5ab811d126ed8fff8fd9dc43dbbb845ee Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 20:46:52 +0200
Subject: [PATCH 150/267] sched-no-work-when-pi-blocked.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/sched.h | 8 ++++++++
kernel/sched.c | 5 ++++-
2 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 38d78dc..99e7f8b 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2095,12 +2095,20 @@ extern unsigned int sysctl_sched_cfs_bandwidth_slice;
extern int rt_mutex_getprio(struct task_struct *p);
extern void rt_mutex_setprio(struct task_struct *p, int prio);
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;
}
# define rt_mutex_adjust_pi(p) do { } while (0)
+static inline bool tsk_is_pi_blocked(struct task_struct *tsk)
+{
+ return false;
+}
#endif
extern bool yield_to(struct task_struct *p, bool preempt);
diff --git a/kernel/sched.c b/kernel/sched.c
index af3883b..5b871cb 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4461,7 +4461,7 @@ need_resched:
static inline void sched_submit_work(struct task_struct *tsk)
{
- if (!tsk->state)
+ if (!tsk->state || tsk_is_pi_blocked(tsk))
return;
/*
@@ -4481,6 +4481,9 @@ static inline void sched_submit_work(struct task_struct *tsk)
static inline void sched_update_worker(struct task_struct *tsk)
{
+ if (tsk_is_pi_blocked(tsk))
+ return;
+
if (tsk->flags & PF_WQ_WORKER)
wq_worker_running(tsk);
}
--
1.7.10

33
debian/patches/features/all/rt/0152-sched-disable-ttwu-queue.patch.patch vendored
View File

@ -1,33 +0,0 @@
From 0366f3687cfda21ddb7b00c56165439afb0a2ffb Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 13 Sep 2011 16:42:35 +0200
Subject: [PATCH 152/267] sched-disable-ttwu-queue.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched_features.h | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/kernel/sched_features.h b/kernel/sched_features.h
index 8480224..0007001 100644
--- a/kernel/sched_features.h
+++ b/kernel/sched_features.h
@@ -60,11 +60,15 @@ SCHED_FEAT(OWNER_SPIN, 1)
*/
SCHED_FEAT(NONTASK_POWER, 1)
+#ifndef CONFIG_PREEMPT_RT_FULL
/*
* Queue remote wakeups on the target CPU and process them
* using the scheduler IPI. Reduces rq->lock contention/bounces.
*/
SCHED_FEAT(TTWU_QUEUE, 1)
+#else
+SCHED_FEAT(TTWU_QUEUE, 0)
+#endif
SCHED_FEAT(FORCE_SD_OVERLAP, 0)
SCHED_FEAT(RT_RUNTIME_SHARE, 1)
--
1.7.10

31
debian/patches/features/all/rt/0174-sched-clear-pf-thread-bound-on-fallback-rq.patch.patch vendored
View File

@ -1,31 +0,0 @@
From 15761075fe95848ac4cfa95a0b2e860eb2e3b82a Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 4 Nov 2011 20:48:36 +0100
Subject: [PATCH 174/267] sched-clear-pf-thread-bound-on-fallback-rq.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/sched.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/kernel/sched.c b/kernel/sched.c
index 51c0ad6..81b340d 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2570,7 +2570,12 @@ static int select_fallback_rq(int cpu, struct task_struct *p)
printk(KERN_INFO "process %d (%s) no longer affine to cpu%d\n",
task_pid_nr(p), p->comm, cpu);
}
-
+ /*
+ * Clear PF_THREAD_BOUND, otherwise we wreckage
+ * migrate_disable/enable. See optimization for
+ * PF_THREAD_BOUND tasks there.
+ */
+ p->flags &= ~PF_THREAD_BOUND;
return dest_cpu;
}
--
1.7.10

47
debian/patches/features/all/rt/0211-serial-8250-Clean-up-the-locking-for-rt.patch vendored
View File

@ -1,47 +0,0 @@
From 40ebaf9abcb5b1d219eba80ba6648c0142b9fa12 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:30:01 -0500
Subject: [PATCH 211/267] serial: 8250: Clean up the locking for -rt
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/tty/serial/8250.c | 15 +++++----------
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/drivers/tty/serial/8250.c b/drivers/tty/serial/8250.c
index eeadf1b..43f7f46 100644
--- a/drivers/tty/serial/8250.c
+++ b/drivers/tty/serial/8250.c
@@ -2846,14 +2846,10 @@ serial8250_console_write(struct console *co, const char *s, unsigned int count)
touch_nmi_watchdog();
- local_irq_save(flags);
- if (up->port.sysrq) {
- /* serial8250_handle_port() already took the lock */
- locked = 0;
- } else if (oops_in_progress) {
- locked = spin_trylock(&up->port.lock);
- } else
- spin_lock(&up->port.lock);
+ if (up->port.sysrq || oops_in_progress)
+ locked = spin_trylock_irqsave(&up->port.lock, flags);
+ else
+ spin_lock_irqsave(&up->port.lock, flags);
/*
* First save the IER then disable the interrupts
@@ -2885,8 +2881,7 @@ serial8250_console_write(struct console *co, const char *s, unsigned int count)
check_modem_status(up);
if (locked)
- spin_unlock(&up->port.lock);
- local_irq_restore(flags);
+ spin_unlock_irqrestore(&up->port.lock, flags);
}
static int __init serial8250_console_setup(struct console *co, char *options)
--
1.7.10

43
debian/patches/features/all/rt/0213-drivers-tty-fix-omap-lock-crap.patch.patch vendored
View File

@ -1,43 +0,0 @@
From 81272c044fe3dd9a13be10c4aca5a0d6d6d428e2 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 28 Jul 2011 13:32:57 +0200
Subject: [PATCH 213/267] drivers-tty-fix-omap-lock-crap.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/tty/serial/omap-serial.c | 8 +++-----
1 file changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
index 5e713d3..93cdb92 100644
--- a/drivers/tty/serial/omap-serial.c
+++ b/drivers/tty/serial/omap-serial.c
@@ -946,13 +946,12 @@ serial_omap_console_write(struct console *co, const char *s,
unsigned int ier;
int locked = 1;
- local_irq_save(flags);
if (up->port.sysrq)
locked = 0;
else if (oops_in_progress)
- locked = spin_trylock(&up->port.lock);
+ locked = spin_trylock_irqsave(&up->port.lock, flags);
else
- spin_lock(&up->port.lock);
+ spin_lock_irqsave(&up->port.lock, flags);
/*
* First save the IER then disable the interrupts
@@ -979,8 +978,7 @@ serial_omap_console_write(struct console *co, const char *s,
check_modem_status(up);
if (locked)
- spin_unlock(&up->port.lock);
- local_irq_restore(flags);
+ spin_unlock_irqrestore(&up->port.lock, flags);
}
static int __init
--
1.7.10

48
debian/patches/features/all/rt/0215-fs-namespace-preemption-fix.patch vendored
View File

@ -1,48 +0,0 @@
From d9e6aef724ad0444ca59f536062b3ce3755ab8ff Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 19 Jul 2009 08:44:27 -0500
Subject: [PATCH 215/267] fs: namespace preemption fix
On RT we cannot loop with preemption disabled here as
mnt_make_readonly() might have been preempted. We can safely enable
preemption while waiting for MNT_WRITE_HOLD to be cleared. Safe on !RT
as well.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
fs/namespace.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/fs/namespace.c b/fs/namespace.c
index cfc6d44..63fd6d7 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -341,8 +341,14 @@ int mnt_want_write(struct vfsmount *mnt)
* incremented count after it has set MNT_WRITE_HOLD.
*/
smp_mb();
- while (mnt->mnt_flags & MNT_WRITE_HOLD)
+ /*
+ * No need to keep preemption disabled accross the spin loop.
+ */
+ while (mnt->mnt_flags & MNT_WRITE_HOLD) {
+ preempt_enable();
cpu_relax();
+ preempt_disable();
+ }
/*
* After the slowpath clears MNT_WRITE_HOLD, mnt_is_readonly will
* be set to match its requirements. So we must not load that until
@@ -352,9 +358,7 @@ int mnt_want_write(struct vfsmount *mnt)
if (__mnt_is_readonly(mnt)) {
mnt_dec_writers(mnt);
ret = -EROFS;
- goto out;
}
-out:
preempt_enable();
return ret;
}
--
1.7.10

26
debian/patches/features/all/rt/0231-jump-label-rt.patch.patch vendored
View File

@ -1,26 +0,0 @@
From 2bd1f5a798104f32a4dd4dff8fcf06c97cbba2ba Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 13 Jul 2011 11:03:16 +0200
Subject: [PATCH 231/267] jump-label-rt.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/jump_label.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h
index 388b0d4..9cc8ed9 100644
--- a/include/linux/jump_label.h
+++ b/include/linux/jump_label.h
@@ -4,7 +4,7 @@
#include <linux/types.h>
#include <linux/compiler.h>
-#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_JUMP_LABEL)
+#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_JUMP_LABEL) && !defined(CONFIG_PREEMPT_BASE)
struct jump_label_key {
atomic_t enabled;
--
1.7.10

64
debian/patches/features/all/rt/0235-printk-Disable-migration-instead-of-preemption.patch vendored
View File

@ -1,64 +0,0 @@
From 2393db19529654fe543796abc967c908eb7c6148 Mon Sep 17 00:00:00 2001
From: Richard Weinberger <rw@linutronix.de>
Date: Mon, 12 Dec 2011 14:35:56 +0100
Subject: [PATCH 235/267] printk: Disable migration instead of preemption
There is no need do disable preemption in vprintk(), disable_migrate()
is sufficient. This fixes the following bug in -rt:
[ 14.759233] BUG: sleeping function called from invalid context
at /home/rw/linux-rt/kernel/rtmutex.c:645
[ 14.759235] in_atomic(): 1, irqs_disabled(): 0, pid: 547, name: bash
[ 14.759244] Pid: 547, comm: bash Not tainted 3.0.12-rt29+ #3
[ 14.759246] Call Trace:
[ 14.759301] [<ffffffff8106fade>] __might_sleep+0xeb/0xf0
[ 14.759318] [<ffffffff810ad784>] rt_spin_lock_fastlock.constprop.9+0x21/0x43
[ 14.759336] [<ffffffff8161fef0>] rt_spin_lock+0xe/0x10
[ 14.759354] [<ffffffff81347ad1>] serial8250_console_write+0x81/0x121
[ 14.759366] [<ffffffff8107ecd3>] __call_console_drivers+0x7c/0x93
[ 14.759369] [<ffffffff8107ef31>] _call_console_drivers+0x5c/0x60
[ 14.759372] [<ffffffff8107f7e5>] console_unlock+0x147/0x1a2
[ 14.759374] [<ffffffff8107fd33>] vprintk+0x3ea/0x462
[ 14.759383] [<ffffffff816160e0>] printk+0x51/0x53
[ 14.759399] [<ffffffff811974e4>] ? proc_reg_poll+0x9a/0x9a
[ 14.759403] [<ffffffff81335b42>] __handle_sysrq+0x50/0x14d
[ 14.759406] [<ffffffff81335c8a>] write_sysrq_trigger+0x4b/0x53
[ 14.759408] [<ffffffff81335c3f>] ? __handle_sysrq+0x14d/0x14d
[ 14.759410] [<ffffffff81197583>] proc_reg_write+0x9f/0xbe
[ 14.759426] [<ffffffff811497ec>] vfs_write+0xac/0xf3
[ 14.759429] [<ffffffff8114a9b3>] ? fget_light+0x3a/0x9b
[ 14.759431] [<ffffffff811499db>] sys_write+0x4a/0x6e
[ 14.759438] [<ffffffff81625d52>] system_call_fastpath+0x16/0x1b
Signed-off-by: Richard Weinberger <rw@linutronix.de>
Link: http://lkml.kernel.org/r/1323696956-11445-1-git-send-email-rw@linutronix.de
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/printk.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/kernel/printk.c b/kernel/printk.c
index fad23c8..13ea6a9 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -901,7 +901,7 @@ asmlinkage int vprintk(const char *fmt, va_list args)
boot_delay_msec();
printk_delay();
- preempt_disable();
+ migrate_disable();
/* This stops the holder of console_sem just where we want him */
raw_local_irq_save(flags);
this_cpu = smp_processor_id();
@@ -1032,7 +1032,7 @@ asmlinkage int vprintk(const char *fmt, va_list args)
out_restore_irqs:
raw_local_irq_restore(flags);
- preempt_enable();
+ migrate_enable();
return printed_len;
}
EXPORT_SYMBOL(printk);
--
1.7.10

169
debian/patches/features/all/rt/0247-sysrq-Allow-immediate-Magic-SysRq-output-for-PREEMPT.patch vendored
View File

@ -1,169 +0,0 @@
From 5f911928368b0d9b62da0df476c458037b4f7da4 Mon Sep 17 00:00:00 2001
From: Frank Rowand <frank.rowand@am.sony.com>
Date: Fri, 23 Sep 2011 13:43:12 -0700
Subject: [PATCH 247/267] sysrq: Allow immediate Magic SysRq output for
PREEMPT_RT_FULL
Add a CONFIG option to allow the output from Magic SysRq to be output
immediately, even if this causes large latencies.
If PREEMPT_RT_FULL, printk() will not try to acquire the console lock
when interrupts or preemption are disabled. If the console lock is
not acquired the printk() output will be buffered, but will not be
output immediately. Some drivers call into the Magic SysRq code
with interrupts or preemption disabled, so the output of Magic SysRq
will be buffered instead of printing immediately if this option is
not selected.
Even with this option selected, Magic SysRq output will be delayed
if the attempt to acquire the console lock fails.
Signed-off-by: Frank Rowand <frank.rowand@am.sony.com>
Link: http://lkml.kernel.org/r/4E7CEF60.5020508@am.sony.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
[ukleinek: make apply on top of debian/sysrq-mask.patch]
drivers/tty/serial/cpm_uart/cpm_uart_core.c | 2 +-
drivers/tty/sysrq.c | 23 +++++++++++++++++++++++
include/linux/sysrq.h | 5 +++++
kernel/printk.c | 5 +++--
lib/Kconfig.debug | 22 ++++++++++++++++++++++
5 files changed, 54 insertions(+), 3 deletions(-)
diff --git a/drivers/tty/serial/cpm_uart/cpm_uart_core.c b/drivers/tty/serial/cpm_uart/cpm_uart_core.c
index b418947..a8b0559 100644
--- a/drivers/tty/serial/cpm_uart/cpm_uart_core.c
+++ b/drivers/tty/serial/cpm_uart/cpm_uart_core.c
@@ -1226,7 +1226,7 @@ static void cpm_uart_console_write(struct console *co, const char *s,
{
struct uart_cpm_port *pinfo = &cpm_uart_ports[co->index];
unsigned long flags;
- int nolock = oops_in_progress;
+ int nolock = oops_in_progress || sysrq_in_progress;
if (unlikely(nolock)) {
local_irq_save(flags);
diff --git a/drivers/tty/sysrq.c b/drivers/tty/sysrq.c
index 43db715..5219738 100644
--- a/drivers/tty/sysrq.c
+++ b/drivers/tty/sysrq.c
@@ -492,6 +492,23 @@ static void __sysrq_put_key_op(int key, struct sysrq_key_op *op_p)
sysrq_key_table[i] = op_p;
}
+#ifdef CONFIG_MAGIC_SYSRQ_FORCE_PRINTK
+
+int sysrq_in_progress;
+
+static void set_sysrq_in_progress(int value)
+{
+ sysrq_in_progress = value;
+}
+
+#else
+
+static void set_sysrq_in_progress(int value)
+{
+}
+
+#endif
+
void __handle_sysrq(int key, bool check_mask)
{
struct sysrq_key_op *op_p;
@@ -500,6 +517,9 @@ void __handle_sysrq(int key, bool check_mask)
unsigned long flags;
spin_lock_irqsave(&sysrq_key_table_lock, flags);
+
+ set_sysrq_in_progress(1);
+
/*
* Raise the apparent loglevel to maximum so that the sysrq header
* is shown to provide the user with positive feedback. We do not
@@ -541,6 +561,9 @@ void __handle_sysrq(int key, bool check_mask)
printk("\n");
console_loglevel = orig_log_level;
}
+
+ set_sysrq_in_progress(0);
+
spin_unlock_irqrestore(&sysrq_key_table_lock, flags);
}
diff --git a/include/linux/sysrq.h b/include/linux/sysrq.h
index 7faf933..d224c0b 100644
--- a/include/linux/sysrq.h
+++ b/include/linux/sysrq.h
@@ -38,6 +38,11 @@ struct sysrq_key_op {
int enable_mask;
};
+#ifdef CONFIG_MAGIC_SYSRQ_FORCE_PRINTK
+extern int sysrq_in_progress;
+#else
+#define sysrq_in_progress 0
+#endif
#ifdef CONFIG_MAGIC_SYSRQ
/* Generic SysRq interface -- you may call it from any device driver, supplying
diff --git a/kernel/printk.c b/kernel/printk.c
index 13ea6a9..9eabbbb 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -21,6 +21,7 @@
#include <linux/tty.h>
#include <linux/tty_driver.h>
#include <linux/console.h>
+#include <linux/sysrq.h>
#include <linux/init.h>
#include <linux/jiffies.h>
#include <linux/nmi.h>
@@ -834,8 +835,8 @@ static int console_trylock_for_printk(unsigned int cpu, unsigned long flags)
{
int retval = 0, wake = 0;
#ifdef CONFIG_PREEMPT_RT_FULL
- int lock = !early_boot_irqs_disabled && !irqs_disabled_flags(flags) &&
- !preempt_count();
+ int lock = (!early_boot_irqs_disabled && !irqs_disabled_flags(flags) &&
+ !preempt_count()) || sysrq_in_progress;
#else
int lock = 1;
#endif
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index c347db3..13a937b 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -62,6 +62,28 @@ config MAGIC_SYSRQ
Specifies the default mask for the allowed SysRq keys. This can be
used to disable several sensitive keys by default.
+config MAGIC_SYSRQ_FORCE_PRINTK
+ bool "Force printk from Magic SysRq"
+ depends on MAGIC_SYSRQ && PREEMPT_RT_FULL
+ default n
+ help
+ Allow the output from Magic SysRq to be output immediately, even if
+ this causes large latencies. This can cause performance problems
+ for real-time processes.
+
+ If PREEMPT_RT_FULL, printk() will not try to acquire the console lock
+ when interrupts or preemption are disabled. If the console lock is
+ not acquired the printk() output will be buffered, but will not be
+ output immediately. Some drivers call into the Magic SysRq code
+ with interrupts or preemption disabled, so the output of Magic SysRq
+ will be buffered instead of printing immediately if this option is
+ not selected.
+
+ Even with this option selected, Magic SysRq output will be delayed
+ if the attempt to acquire the console lock fails.
+
+ Don't say Y unless you really know what this hack does.
+
config STRIP_ASM_SYMS
bool "Strip assembler-generated symbols during link"
default n
--
1.7.10

158
debian/patches/features/all/rt/0260-softirq-Check-preemption-after-reenabling-interrupts.patch vendored
View File

@ -1,158 +0,0 @@
From a583b75cab964e833c5031f86eb95d3caed27398 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 13 Nov 2011 17:17:09 +0100
Subject: [PATCH 260/267] softirq: Check preemption after reenabling
interrupts
raise_softirq_irqoff() disables interrupts and wakes the softirq
daemon, but after reenabling interrupts there is no preemption check,
so the execution of the softirq thread might be delayed arbitrarily.
In principle we could add that check to local_irq_enable/restore, but
that's overkill as the rasie_softirq_irqoff() sections are the only
ones which show this behaviour.
Reported-by: Carsten Emde <cbe@osadl.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable-rt@vger.kernel.org
---
block/blk-iopoll.c | 3 +++
block/blk-softirq.c | 3 +++
include/linux/preempt.h | 3 +++
net/core/dev.c | 6 ++++++
4 files changed, 15 insertions(+)
diff --git a/block/blk-iopoll.c b/block/blk-iopoll.c
index 58916af..f7ca9b4 100644
--- a/block/blk-iopoll.c
+++ b/block/blk-iopoll.c
@@ -38,6 +38,7 @@ void blk_iopoll_sched(struct blk_iopoll *iop)
list_add_tail(&iop->list, &__get_cpu_var(blk_cpu_iopoll));
__raise_softirq_irqoff(BLOCK_IOPOLL_SOFTIRQ);
local_irq_restore(flags);
+ preempt_check_resched_rt();
}
EXPORT_SYMBOL(blk_iopoll_sched);
@@ -135,6 +136,7 @@ static void blk_iopoll_softirq(struct softirq_action *h)
__raise_softirq_irqoff(BLOCK_IOPOLL_SOFTIRQ);
local_irq_enable();
+ preempt_check_resched_rt();
}
/**
@@ -204,6 +206,7 @@ static int __cpuinit blk_iopoll_cpu_notify(struct notifier_block *self,
&__get_cpu_var(blk_cpu_iopoll));
__raise_softirq_irqoff(BLOCK_IOPOLL_SOFTIRQ);
local_irq_enable();
+ preempt_check_resched_rt();
}
return NOTIFY_OK;
diff --git a/block/blk-softirq.c b/block/blk-softirq.c
index 1366a89..60a88ab 100644
--- a/block/blk-softirq.c
+++ b/block/blk-softirq.c
@@ -50,6 +50,7 @@ static void trigger_softirq(void *data)
raise_softirq_irqoff(BLOCK_SOFTIRQ);
local_irq_restore(flags);
+ preempt_check_resched_rt();
}
/*
@@ -92,6 +93,7 @@ static int __cpuinit blk_cpu_notify(struct notifier_block *self,
&__get_cpu_var(blk_cpu_done));
raise_softirq_irqoff(BLOCK_SOFTIRQ);
local_irq_enable();
+ preempt_check_resched_rt();
}
return NOTIFY_OK;
@@ -150,6 +152,7 @@ do_local:
goto do_local;
local_irq_restore(flags);
+ preempt_check_resched_rt();
}
/**
diff --git a/include/linux/preempt.h b/include/linux/preempt.h
index 6450c01..58d8982 100644
--- a/include/linux/preempt.h
+++ b/include/linux/preempt.h
@@ -56,8 +56,10 @@ do { \
#ifndef CONFIG_PREEMPT_RT_BASE
# define preempt_enable_no_resched() __preempt_enable_no_resched()
+# define preempt_check_resched_rt() do { } while (0)
#else
# define preempt_enable_no_resched() preempt_enable()
+# define preempt_check_resched_rt() preempt_check_resched()
#endif
#define preempt_enable() \
@@ -105,6 +107,7 @@ do { \
#define preempt_disable_notrace() do { } while (0)
#define preempt_enable_no_resched_notrace() do { } while (0)
#define preempt_enable_notrace() do { } while (0)
+#define preempt_check_resched_rt() do { } while (0)
#endif /* CONFIG_PREEMPT_COUNT */
diff --git a/net/core/dev.c b/net/core/dev.c
index ae4fd5e..3756e96 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1779,6 +1779,7 @@ static inline void __netif_reschedule(struct Qdisc *q)
sd->output_queue_tailp = &q->next_sched;
raise_softirq_irqoff(NET_TX_SOFTIRQ);
local_irq_restore(flags);
+ preempt_check_resched_rt();
}
void __netif_schedule(struct Qdisc *q)
@@ -1800,6 +1801,7 @@ void dev_kfree_skb_irq(struct sk_buff *skb)
sd->completion_queue = skb;
raise_softirq_irqoff(NET_TX_SOFTIRQ);
local_irq_restore(flags);
+ preempt_check_resched_rt();
}
}
EXPORT_SYMBOL(dev_kfree_skb_irq);
@@ -2969,6 +2971,7 @@ enqueue:
rps_unlock(sd);
local_irq_restore(flags);
+ preempt_check_resched_rt();
atomic_long_inc(&skb->dev->rx_dropped);
kfree_skb(skb);
@@ -3790,6 +3793,7 @@ static void net_rps_action_and_irq_enable(struct softnet_data *sd)
} else
#endif
local_irq_enable();
+ preempt_check_resched_rt();
}
static int process_backlog(struct napi_struct *napi, int quota)
@@ -3862,6 +3866,7 @@ void __napi_schedule(struct napi_struct *n)
local_irq_save(flags);
____napi_schedule(&__get_cpu_var(softnet_data), n);
local_irq_restore(flags);
+ preempt_check_resched_rt();
}
EXPORT_SYMBOL(__napi_schedule);
@@ -6402,6 +6407,7 @@ static int dev_cpu_callback(struct notifier_block *nfb,
raise_softirq_irqoff(NET_TX_SOFTIRQ);
local_irq_enable();
+ preempt_check_resched_rt();
/* Process offline CPU's input_pkt_queue */
while ((skb = __skb_dequeue(&oldsd->process_queue))) {
--
1.7.10

54
debian/patches/features/all/rt/0266-rt-Make-migrate_disable-enable-and-__rt_mutex_init-n.patch vendored
View File

@ -1,54 +0,0 @@
From 87b9564badf043711b03d381536be29103e9cce9 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 16 Apr 2012 21:51:54 -0400
Subject: [PATCH 266/267] rt: Make migrate_disable/enable() and
__rt_mutex_init non-GPL only
Modules that load on the normal vanilla kernel should also load on
an -rt kernel as well. This does not mean we condone non-GPL modules,
we are only being consistent.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
---
kernel/rtmutex.c | 2 +-
kernel/sched.c | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c
index b525158..9c4f6e5 100644
--- a/kernel/rtmutex.c
+++ b/kernel/rtmutex.c
@@ -1290,7 +1290,7 @@ void __rt_mutex_init(struct rt_mutex *lock, const char *name)
debug_rt_mutex_init(lock, name);
}
-EXPORT_SYMBOL_GPL(__rt_mutex_init);
+EXPORT_SYMBOL(__rt_mutex_init);
/**
* rt_mutex_init_proxy_locked - initialize and lock a rt_mutex on behalf of a
diff --git a/kernel/sched.c b/kernel/sched.c
index 1cc706d..3b12e0b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4426,7 +4426,7 @@ void migrate_disable(void)
p->migrate_disable = 1;
preempt_enable();
}
-EXPORT_SYMBOL_GPL(migrate_disable);
+EXPORT_SYMBOL(migrate_disable);
void migrate_enable(void)
{
@@ -4478,7 +4478,7 @@ void migrate_enable(void)
unpin_current_cpu();
preempt_enable();
}
-EXPORT_SYMBOL_GPL(migrate_enable);
+EXPORT_SYMBOL(migrate_enable);
#else
static inline void update_migrate_disable(struct task_struct *p) { }
#define migrate_disabled_updated(p) 0
--
1.7.10

19
debian/patches/features/all/rt/0267-Linux-3.2.16-rt27-REBASE.patch vendored
View File

@ -1,19 +0,0 @@
From 9eee911c32b17fa49c4d00541a7459ab912d4fc5 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Mon, 16 Apr 2012 22:15:29 -0400
Subject: [PATCH 267/267] Linux 3.2.16-rt27 REBASE
---
localversion-rt | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/localversion-rt b/localversion-rt
index b2111a2..be1e37b 100644
--- a/localversion-rt
+++ b/localversion-rt
@@ -1 +1 @@
--rt24
+-rt27
--
1.7.10

15
debian/patches/features/all/rt/0084-acpi-Do-not-disable-interrupts-on-PREEMPT_RT.patch → debian/patches/features/all/rt/acpi-use-local-irq-nort.patch vendored
View File

@ -1,19 +1,19 @@
From 6bd3fc45f9c1bd877a0fd6776610ca5658dae3b1 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 21 Jul 2009 22:54:51 +0200
Subject: [PATCH 084/267] acpi: Do not disable interrupts on PREEMPT_RT
Subject: acpi: Do not disable interrupts on PREEMPT_RT
Use the local_irq_*_nort() variants.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/include/asm/acpi.h | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h
index 610001d..c1c23d2 100644
--- a/arch/x86/include/asm/acpi.h
+++ b/arch/x86/include/asm/acpi.h
Index: linux-3.4/arch/x86/include/asm/acpi.h
===================================================================
--- linux-3.4.orig/arch/x86/include/asm/acpi.h
+++ linux-3.4/arch/x86/include/asm/acpi.h
@@ -51,8 +51,8 @@
#define ACPI_ASM_MACROS
@ -25,6 +25,3 @@ index 610001d..c1c23d2 100644
#define ACPI_FLUSH_CPU_CACHE() wbinvd()
int __acpi_acquire_global_lock(unsigned int *lock);
--
1.7.10

241
debian/patches/features/all/rt/0046-mm-Fixup-all-fault-handlers-to-check-current-pagefau.patch → debian/patches/features/all/rt/arch-use-pagefault-disabled.patch vendored
View File

@ -1,8 +1,6 @@
From ce05fde6b2b86bc5569433eef6869bdb1adcb32c Mon Sep 17 00:00:00 2001
Subject: mm: Fixup all fault handlers to check current->pagefault_disable
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 17 Mar 2011 11:32:28 +0100
Subject: [PATCH 046/267] mm: Fixup all fault handlers to check
current->pagefault_disable
Necessary for decoupling pagefault disable from preempt count.
@ -32,11 +30,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arch/xtensa/mm/fault.c | 2 +-
22 files changed, 27 insertions(+), 24 deletions(-)
diff --git a/arch/alpha/mm/fault.c b/arch/alpha/mm/fault.c
index fadd5f8..6d73e1b 100644
--- a/arch/alpha/mm/fault.c
+++ b/arch/alpha/mm/fault.c
@@ -107,7 +107,7 @@ do_page_fault(unsigned long address, unsigned long mmcsr,
Index: linux-3.4/arch/alpha/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/alpha/mm/fault.c
+++ linux-3.4/arch/alpha/mm/fault.c
@@ -106,7 +106,7 @@ do_page_fault(unsigned long address, uns
/* If we're in an interrupt context, or have no user context,
we must not take the fault. */
@ -45,11 +43,11 @@ index fadd5f8..6d73e1b 100644
goto no_context;
#ifdef CONFIG_ALPHA_LARGE_VMALLOC
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index aa33949..c8dd303 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -294,7 +294,7 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
Index: linux-3.4/arch/arm/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/arm/mm/fault.c
+++ linux-3.4/arch/arm/mm/fault.c
@@ -277,7 +277,7 @@ do_page_fault(unsigned long addr, unsign
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -58,11 +56,11 @@ index aa33949..c8dd303 100644
goto no_context;
/*
diff --git a/arch/avr32/mm/fault.c b/arch/avr32/mm/fault.c
index f7040a1..623a027 100644
--- a/arch/avr32/mm/fault.c
+++ b/arch/avr32/mm/fault.c
@@ -81,7 +81,8 @@ asmlinkage void do_page_fault(unsigned long ecr, struct pt_regs *regs)
Index: linux-3.4/arch/avr32/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/avr32/mm/fault.c
+++ linux-3.4/arch/avr32/mm/fault.c
@@ -81,7 +81,8 @@ asmlinkage void do_page_fault(unsigned l
* If we're in an interrupt or have no user context, we must
* not take the fault...
*/
@ -72,11 +70,11 @@ index f7040a1..623a027 100644
goto no_context;
local_irq_enable();
diff --git a/arch/cris/mm/fault.c b/arch/cris/mm/fault.c
index 9dcac8e..2b2c292 100644
--- a/arch/cris/mm/fault.c
+++ b/arch/cris/mm/fault.c
@@ -111,7 +111,7 @@ do_page_fault(unsigned long address, struct pt_regs *regs,
Index: linux-3.4/arch/cris/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/cris/mm/fault.c
+++ linux-3.4/arch/cris/mm/fault.c
@@ -112,7 +112,7 @@ do_page_fault(unsigned long address, str
* user context, we must not take the fault.
*/
@ -85,11 +83,11 @@ index 9dcac8e..2b2c292 100644
goto no_context;
down_read(&mm->mmap_sem);
diff --git a/arch/frv/mm/fault.c b/arch/frv/mm/fault.c
index a325d57..3da8ec7 100644
--- a/arch/frv/mm/fault.c
+++ b/arch/frv/mm/fault.c
@@ -79,7 +79,7 @@ asmlinkage void do_page_fault(int datammu, unsigned long esr0, unsigned long ear
Index: linux-3.4/arch/frv/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/frv/mm/fault.c
+++ linux-3.4/arch/frv/mm/fault.c
@@ -78,7 +78,7 @@ asmlinkage void do_page_fault(int datamm
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -98,11 +96,11 @@ index a325d57..3da8ec7 100644
goto no_context;
down_read(&mm->mmap_sem);
diff --git a/arch/ia64/mm/fault.c b/arch/ia64/mm/fault.c
index 20b3593..2a4e44f 100644
--- a/arch/ia64/mm/fault.c
+++ b/arch/ia64/mm/fault.c
@@ -89,7 +89,7 @@ ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *re
Index: linux-3.4/arch/ia64/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/ia64/mm/fault.c
+++ linux-3.4/arch/ia64/mm/fault.c
@@ -88,7 +88,7 @@ ia64_do_page_fault (unsigned long addres
/*
* If we're in an interrupt or have no user context, we must not take the fault..
*/
@ -111,11 +109,11 @@ index 20b3593..2a4e44f 100644
goto no_context;
#ifdef CONFIG_VIRTUAL_MEM_MAP
diff --git a/arch/m32r/mm/fault.c b/arch/m32r/mm/fault.c
index 2c9aeb4..16fa2c7 100644
--- a/arch/m32r/mm/fault.c
+++ b/arch/m32r/mm/fault.c
@@ -115,7 +115,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long error_code,
Index: linux-3.4/arch/m32r/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/m32r/mm/fault.c
+++ linux-3.4/arch/m32r/mm/fault.c
@@ -114,7 +114,7 @@ asmlinkage void do_page_fault(struct pt_
* If we're in an interrupt or have no user context or are running in an
* atomic region then we must not take the fault..
*/
@ -124,11 +122,11 @@ index 2c9aeb4..16fa2c7 100644
goto bad_area_nosemaphore;
/* When running in the kernel we expect faults to occur only to
diff --git a/arch/m68k/mm/fault.c b/arch/m68k/mm/fault.c
index 2db6099..238ffc0 100644
--- a/arch/m68k/mm/fault.c
+++ b/arch/m68k/mm/fault.c
@@ -85,7 +85,7 @@ int do_page_fault(struct pt_regs *regs, unsigned long address,
Index: linux-3.4/arch/m68k/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/m68k/mm/fault.c
+++ linux-3.4/arch/m68k/mm/fault.c
@@ -84,7 +84,7 @@ int do_page_fault(struct pt_regs *regs,
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -137,11 +135,11 @@ index 2db6099..238ffc0 100644
goto no_context;
down_read(&mm->mmap_sem);
diff --git a/arch/microblaze/mm/fault.c b/arch/microblaze/mm/fault.c
index ae97d2c..c3f219c 100644
--- a/arch/microblaze/mm/fault.c
+++ b/arch/microblaze/mm/fault.c
@@ -107,7 +107,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long address,
Index: linux-3.4/arch/microblaze/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/microblaze/mm/fault.c
+++ linux-3.4/arch/microblaze/mm/fault.c
@@ -106,7 +106,7 @@ void do_page_fault(struct pt_regs *regs,
if ((error_code & 0x13) == 0x13 || (error_code & 0x11) == 0x11)
is_write = 0;
@ -150,11 +148,11 @@ index ae97d2c..c3f219c 100644
if (kernel_mode(regs))
goto bad_area_nosemaphore;
diff --git a/arch/mips/mm/fault.c b/arch/mips/mm/fault.c
index 937cf33..ce7e75e 100644
--- a/arch/mips/mm/fault.c
+++ b/arch/mips/mm/fault.c
@@ -88,7 +88,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs, unsigned long writ
Index: linux-3.4/arch/mips/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/mips/mm/fault.c
+++ linux-3.4/arch/mips/mm/fault.c
@@ -89,7 +89,7 @@ asmlinkage void __kprobes do_page_fault(
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -162,12 +160,12 @@ index 937cf33..ce7e75e 100644
+ if (in_atomic() || !mm || current->pagefault_disabled)
goto bad_area_nosemaphore;
down_read(&mm->mmap_sem);
diff --git a/arch/mn10300/mm/fault.c b/arch/mn10300/mm/fault.c
index 0945409..53c8d16 100644
--- a/arch/mn10300/mm/fault.c
+++ b/arch/mn10300/mm/fault.c
@@ -168,7 +168,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long fault_code,
retry:
Index: linux-3.4/arch/mn10300/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/mn10300/mm/fault.c
+++ linux-3.4/arch/mn10300/mm/fault.c
@@ -167,7 +167,7 @@ asmlinkage void do_page_fault(struct pt_
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -176,11 +174,11 @@ index 0945409..53c8d16 100644
goto no_context;
down_read(&mm->mmap_sem);
diff --git a/arch/parisc/mm/fault.c b/arch/parisc/mm/fault.c
index 18162ce..09ecc8a 100644
--- a/arch/parisc/mm/fault.c
+++ b/arch/parisc/mm/fault.c
@@ -176,7 +176,7 @@ void do_page_fault(struct pt_regs *regs, unsigned long code,
Index: linux-3.4/arch/parisc/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/parisc/mm/fault.c
+++ linux-3.4/arch/parisc/mm/fault.c
@@ -176,7 +176,7 @@ void do_page_fault(struct pt_regs *regs,
unsigned long acc_type;
int fault;
@ -189,24 +187,24 @@ index 18162ce..09ecc8a 100644
goto no_context;
down_read(&mm->mmap_sem);
diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
index 5efe8c9..17f8bbe 100644
--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -162,7 +162,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
}
#endif
Index: linux-3.4/arch/powerpc/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/powerpc/mm/fault.c
+++ linux-3.4/arch/powerpc/mm/fault.c
@@ -261,7 +261,7 @@ int __kprobes do_page_fault(struct pt_re
if (!arch_irq_disabled_regs(regs))
local_irq_enable();
- if (in_atomic() || mm == NULL) {
+ if (in_atomic() || mm == NULL || current->pagefault_disabled) {
if (!user_mode(regs))
return SIGSEGV;
/* in_atomic() in user mode is really bad,
diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c
index c7f0fbc..0d60168 100644
--- a/arch/s390/mm/fault.c
+++ b/arch/s390/mm/fault.c
@@ -294,7 +294,8 @@ static inline int do_exception(struct pt_regs *regs, int access,
Index: linux-3.4/arch/s390/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/s390/mm/fault.c
+++ linux-3.4/arch/s390/mm/fault.c
@@ -283,7 +283,8 @@ static inline int do_exception(struct pt
* user context.
*/
fault = VM_FAULT_BADCONTEXT;
@ -216,21 +214,21 @@ index c7f0fbc..0d60168 100644
goto out;
address = trans_exc_code & __FAIL_ADDR_MASK;
@@ -425,7 +426,8 @@ void __kprobes do_asce_exception(struct pt_regs *regs, long pgm_int_code,
struct mm_struct *mm = current->mm;
struct vm_area_struct *vma;
@@ -415,7 +416,8 @@ void __kprobes do_asce_exception(struct
unsigned long trans_exc_code;
trans_exc_code = regs->int_parm_long;
- if (unlikely(!user_space_fault(trans_exc_code) || in_atomic() || !mm))
+ if (unlikely(!user_space_fault(trans_exc_code) || in_atomic() || !mm ||
+ current->pagefault_disabled))
+ current->pagefault_disabled))
goto no_context;
down_read(&mm->mmap_sem);
diff --git a/arch/score/mm/fault.c b/arch/score/mm/fault.c
index 47b600e..4c12824 100644
--- a/arch/score/mm/fault.c
+++ b/arch/score/mm/fault.c
@@ -72,7 +72,7 @@ asmlinkage void do_page_fault(struct pt_regs *regs, unsigned long write,
Index: linux-3.4/arch/score/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/score/mm/fault.c
+++ linux-3.4/arch/score/mm/fault.c
@@ -72,7 +72,7 @@ asmlinkage void do_page_fault(struct pt_
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -239,11 +237,11 @@ index 47b600e..4c12824 100644
goto bad_area_nosemaphore;
down_read(&mm->mmap_sem);
diff --git a/arch/sh/mm/fault_32.c b/arch/sh/mm/fault_32.c
index 7bebd04..a67ac56 100644
--- a/arch/sh/mm/fault_32.c
+++ b/arch/sh/mm/fault_32.c
@@ -166,7 +166,7 @@ asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
Index: linux-3.4/arch/sh/mm/fault_32.c
===================================================================
--- linux-3.4.orig/arch/sh/mm/fault_32.c
+++ linux-3.4/arch/sh/mm/fault_32.c
@@ -166,7 +166,7 @@ asmlinkage void __kprobes do_page_fault(
* If we're in an interrupt, have no user context or are running
* in an atomic region then we must not take the fault:
*/
@ -252,11 +250,11 @@ index 7bebd04..a67ac56 100644
goto no_context;
down_read(&mm->mmap_sem);
diff --git a/arch/sparc/mm/fault_32.c b/arch/sparc/mm/fault_32.c
index 8023fd7..e0742c1 100644
--- a/arch/sparc/mm/fault_32.c
+++ b/arch/sparc/mm/fault_32.c
@@ -247,8 +247,8 @@ asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write,
Index: linux-3.4/arch/sparc/mm/fault_32.c
===================================================================
--- linux-3.4.orig/arch/sparc/mm/fault_32.c
+++ linux-3.4/arch/sparc/mm/fault_32.c
@@ -248,8 +248,8 @@ asmlinkage void do_sparc_fault(struct pt
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -267,11 +265,11 @@ index 8023fd7..e0742c1 100644
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c
index 504c062..9d691a5 100644
--- a/arch/sparc/mm/fault_64.c
+++ b/arch/sparc/mm/fault_64.c
@@ -322,7 +322,7 @@ asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
Index: linux-3.4/arch/sparc/mm/fault_64.c
===================================================================
--- linux-3.4.orig/arch/sparc/mm/fault_64.c
+++ linux-3.4/arch/sparc/mm/fault_64.c
@@ -323,7 +323,7 @@ asmlinkage void __kprobes do_sparc64_fau
* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -280,11 +278,11 @@ index 504c062..9d691a5 100644
goto intr_or_no_mm;
perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, address);
diff --git a/arch/tile/mm/fault.c b/arch/tile/mm/fault.c
index 25b7b90..ecdb016 100644
--- a/arch/tile/mm/fault.c
+++ b/arch/tile/mm/fault.c
@@ -346,7 +346,7 @@ static int handle_page_fault(struct pt_regs *regs,
Index: linux-3.4/arch/tile/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/tile/mm/fault.c
+++ linux-3.4/arch/tile/mm/fault.c
@@ -355,7 +355,7 @@ static int handle_page_fault(struct pt_r
* If we're in an interrupt, have no user context or are running in an
* atomic region then we must not take the fault.
*/
@ -293,11 +291,11 @@ index 25b7b90..ecdb016 100644
vma = NULL; /* happy compiler */
goto bad_area_nosemaphore;
}
diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c
index dafc947..a283400 100644
--- a/arch/um/kernel/trap.c
+++ b/arch/um/kernel/trap.c
@@ -37,7 +37,7 @@ int handle_page_fault(unsigned long address, unsigned long ip,
Index: linux-3.4/arch/um/kernel/trap.c
===================================================================
--- linux-3.4.orig/arch/um/kernel/trap.c
+++ linux-3.4/arch/um/kernel/trap.c
@@ -37,7 +37,7 @@ int handle_page_fault(unsigned long addr
* If the fault was during atomic operation, don't take the fault, just
* fail.
*/
@ -306,11 +304,11 @@ index dafc947..a283400 100644
goto out_nosemaphore;
down_read(&mm->mmap_sem);
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index 5db0490..191015f 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -1084,7 +1084,7 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code)
Index: linux-3.4/arch/x86/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/x86/mm/fault.c
+++ linux-3.4/arch/x86/mm/fault.c
@@ -1094,7 +1094,7 @@ do_page_fault(struct pt_regs *regs, unsi
* If we're in an interrupt, have no user context or are running
* in an atomic region then we must not take the fault:
*/
@ -319,11 +317,11 @@ index 5db0490..191015f 100644
bad_area_nosemaphore(regs, error_code, address);
return;
}
diff --git a/arch/xtensa/mm/fault.c b/arch/xtensa/mm/fault.c
index e367e30..705f483 100644
--- a/arch/xtensa/mm/fault.c
+++ b/arch/xtensa/mm/fault.c
@@ -57,7 +57,7 @@ void do_page_fault(struct pt_regs *regs)
Index: linux-3.4/arch/xtensa/mm/fault.c
===================================================================
--- linux-3.4.orig/arch/xtensa/mm/fault.c
+++ linux-3.4/arch/xtensa/mm/fault.c
@@ -56,7 +56,7 @@ void do_page_fault(struct pt_regs *regs)
/* If we're in an interrupt or have no user
* context, we must not take the fault..
*/
@ -332,6 +330,3 @@ index e367e30..705f483 100644
bad_page_fault(regs, address, SIGSEGV);
return;
}
--
1.7.10

22
debian/patches/features/all/rt/0028-arm-Allow-forced-irq-threading.patch → debian/patches/features/all/rt/arm-allow-irq-threading.patch vendored
View File

@ -1,7 +1,6 @@
From d24eb79d3953d59e4a232bb66d6182d00007c02d Mon Sep 17 00:00:00 2001
Subject: arm: Allow forced irq threading
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 16 Jul 2011 13:15:20 +0200
Subject: [PATCH 028/267] arm: Allow forced irq threading
All timer interrupts and the perf interrupt are marked NO_THREAD, so
its safe to allow forced interrupt threading.
@ -11,18 +10,15 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arch/arm/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index ab3740e..fc327b8 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -29,6 +29,7 @@ config ARM
Index: linux-3.4/arch/arm/Kconfig
===================================================================
--- linux-3.4.orig/arch/arm/Kconfig
+++ linux-3.4/arch/arm/Kconfig
@@ -31,6 +31,7 @@ config ARM
select HAVE_C_RECORDMCOUNT
select HAVE_GENERIC_HARDIRQS
select HAVE_SPARSE_IRQ
select GENERIC_IRQ_SHOW
+ select IRQ_FORCED_THREADING
select CPU_PM if (SUSPEND || CPU_IDLE)
help
The ARM series is a line of low-power-consumption RISC chip designs
--
1.7.10
select GENERIC_PCI_IOMAP
select HAVE_BPF_JIT if NET

58
debian/patches/features/all/rt/arm-at91-pit-remove-irq-handler-when-clock-is-unused.patch vendored Normal file
View File

@ -0,0 +1,58 @@
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
Setup and remove the interrupt handler in clock event mode selection.
This avoids calling the (shared) interrupt handler when the device is
not used.
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/mach-at91/at91rm9200_time.c | 1 +
arch/arm/mach-at91/at91sam926x_time.c | 5 ++++-
2 files changed, 5 insertions(+), 1 deletion(-)
Index: linux-3.4/arch/arm/mach-at91/at91rm9200_time.c
===================================================================
--- linux-3.4.orig/arch/arm/mach-at91/at91rm9200_time.c
+++ linux-3.4/arch/arm/mach-at91/at91rm9200_time.c
@@ -130,6 +130,7 @@ clkevt32k_mode(enum clock_event_mode mod
break;
case CLOCK_EVT_MODE_SHUTDOWN:
case CLOCK_EVT_MODE_UNUSED:
+ remove_irq(AT91_ID_SYS, &at91rm9200_timer_irq);
case CLOCK_EVT_MODE_RESUME:
irqmask = 0;
break;
Index: linux-3.4/arch/arm/mach-at91/at91sam926x_time.c
===================================================================
--- linux-3.4.orig/arch/arm/mach-at91/at91sam926x_time.c
+++ linux-3.4/arch/arm/mach-at91/at91sam926x_time.c
@@ -67,7 +67,7 @@ static struct clocksource pit_clk = {
.flags = CLOCK_SOURCE_IS_CONTINUOUS,
};
-
+static struct irqaction at91sam926x_pit_irq;
/*
* Clockevent device: interrupts every 1/HZ (== pit_cycles * MCK/16)
*/
@@ -76,6 +76,8 @@ pit_clkevt_mode(enum clock_event_mode mo
{
switch (mode) {
case CLOCK_EVT_MODE_PERIODIC:
+ /* Set up irq handler */
+ setup_irq(AT91_ID_SYS, &at91sam926x_pit_irq);
/* update clocksource counter */
pit_cnt += pit_cycle * PIT_PICNT(pit_read(AT91_PIT_PIVR));
pit_write(AT91_PIT_MR, (pit_cycle - 1) | AT91_PIT_PITEN
@@ -88,6 +90,7 @@ pit_clkevt_mode(enum clock_event_mode mo
case CLOCK_EVT_MODE_UNUSED:
/* disable irq, leaving the clocksource active */
pit_write(AT91_PIT_MR, (pit_cycle - 1) | AT91_PIT_PITEN);
+ remove_irq(AT91_ID_SYS, &at91sam926x_pit_irq);
break;
case CLOCK_EVT_MODE_RESUME:
break;

19
debian/patches/features/all/rt/0239-ARM-at91-tclib-Default-to-tclib-timer-for-RT.patch → debian/patches/features/all/rt/arm-at91-tclib-default-to-tclib-timer-for-rt.patch vendored
View File

@ -1,21 +1,21 @@
From 7f2157307f760daa1eb214bbe42f6b128bde99b2 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 1 May 2010 18:29:35 +0200
Subject: [PATCH 239/267] ARM: at91: tclib: Default to tclib timer for RT
Subject: ARM: at91: tclib: Default to tclib timer for RT
RT is not too happy about the shared timer interrupt in AT91
devices. Default to tclib timer for RT.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/misc/Kconfig | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 1cb530c..951ae6c 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -82,6 +82,7 @@ config AB8500_PWM
Index: linux-3.4/drivers/misc/Kconfig
===================================================================
--- linux-3.4.orig/drivers/misc/Kconfig
+++ linux-3.4/drivers/misc/Kconfig
@@ -72,6 +72,7 @@ config AB8500_PWM
config ATMEL_TCLIB
bool "Atmel AT32/AT91 Timer/Counter Library"
depends on (AVR32 || ARCH_AT91)
@ -23,7 +23,7 @@ index 1cb530c..951ae6c 100644
help
Select this if you want a library to allocate the Timer/Counter
blocks found on many Atmel processors. This facilitates using
@@ -114,7 +115,7 @@ config ATMEL_TCB_CLKSRC_BLOCK
@@ -104,7 +105,7 @@ config ATMEL_TCB_CLKSRC_BLOCK
config ATMEL_TCB_CLKSRC_USE_SLOW_CLOCK
bool "TC Block use 32 KiHz clock"
depends on ATMEL_TCB_CLKSRC
@ -32,6 +32,3 @@ index 1cb530c..951ae6c 100644
help
Select this to use 32 KiHz base clock rate as TC block clock
source for clock events.
--
1.7.10

138
debian/patches/features/all/rt/0029-preempt-rt-Convert-arm-boot_lock-to-raw.patch → debian/patches/features/all/rt/arm-convert-boot-lock-to-raw.patch vendored
View File

@ -1,7 +1,7 @@
From b4413229a63e7c48b6f26a03eb74e418dce27d49 Mon Sep 17 00:00:00 2001
Subject: preempt-rt: Convert arm boot_lock to raw
From: Frank Rowand <frank.rowand@am.sony.com>
Date: Mon, 19 Sep 2011 14:51:14 -0700
Subject: [PATCH 029/267] preempt-rt: Convert arm boot_lock to raw
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.
@ -23,25 +23,24 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arch/arm/mach-exynos/platsmp.c | 12 ++++++------
arch/arm/mach-msm/platsmp.c | 10 +++++-----
arch/arm/mach-omap2/omap-smp.c | 10 +++++-----
arch/arm/mach-tegra/platsmp.c | 10 +++++-----
arch/arm/mach-ux500/platsmp.c | 10 +++++-----
arch/arm/plat-versatile/platsmp.c | 10 +++++-----
6 files changed, 31 insertions(+), 31 deletions(-)
5 files changed, 26 insertions(+), 26 deletions(-)
diff --git a/arch/arm/mach-exynos/platsmp.c b/arch/arm/mach-exynos/platsmp.c
index 69ffb2f..fe321b0 100644
--- a/arch/arm/mach-exynos/platsmp.c
+++ b/arch/arm/mach-exynos/platsmp.c
@@ -63,7 +63,7 @@ static void __iomem *scu_base_addr(void)
Index: linux-3.4/arch/arm/mach-exynos/platsmp.c
===================================================================
--- linux-3.4.orig/arch/arm/mach-exynos/platsmp.c
+++ linux-3.4/arch/arm/mach-exynos/platsmp.c
@@ -62,7 +62,7 @@ static void __iomem *scu_base_addr(void)
return (void __iomem *)(S5P_VA_SCU);
}
-static DEFINE_SPINLOCK(boot_lock);
+static DEFINE_RAW_SPINLOCK(boot_lock);
static void __cpuinit exynos4_gic_secondary_init(void)
void __cpuinit platform_secondary_init(unsigned int cpu)
{
@@ -108,8 +108,8 @@ void __cpuinit platform_secondary_init(unsigned int cpu)
@@ -82,8 +82,8 @@ void __cpuinit platform_secondary_init(u
/*
* Synchronise with the boot thread.
*/
@ -52,7 +51,7 @@ index 69ffb2f..fe321b0 100644
}
int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -120,7 +120,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -94,7 +94,7 @@ int __cpuinit boot_secondary(unsigned in
* Set synchronisation state between this boot processor
* and the secondary one
*/
@ -61,7 +60,7 @@ index 69ffb2f..fe321b0 100644
/*
* The secondary processor is waiting to be released from
@@ -149,7 +149,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -123,7 +123,7 @@ int __cpuinit boot_secondary(unsigned in
if (timeout == 0) {
printk(KERN_ERR "cpu1 power enable failed");
@ -70,7 +69,7 @@ index 69ffb2f..fe321b0 100644
return -ETIMEDOUT;
}
}
@@ -177,7 +177,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -151,7 +151,7 @@ int __cpuinit boot_secondary(unsigned in
* now the secondary core is starting up let it run its
* calibrations, then wait for it to finish
*/
@ -79,11 +78,11 @@ index 69ffb2f..fe321b0 100644
return pen_release != -1 ? -ENOSYS : 0;
}
diff --git a/arch/arm/mach-msm/platsmp.c b/arch/arm/mach-msm/platsmp.c
index fdec58a..cad6b81 100644
--- a/arch/arm/mach-msm/platsmp.c
+++ b/arch/arm/mach-msm/platsmp.c
@@ -39,7 +39,7 @@ extern void msm_secondary_startup(void);
Index: linux-3.4/arch/arm/mach-msm/platsmp.c
===================================================================
--- linux-3.4.orig/arch/arm/mach-msm/platsmp.c
+++ linux-3.4/arch/arm/mach-msm/platsmp.c
@@ -40,7 +40,7 @@ extern void msm_secondary_startup(void);
*/
volatile int pen_release = -1;
@ -92,7 +91,7 @@ index fdec58a..cad6b81 100644
static inline int get_core_count(void)
{
@@ -69,8 +69,8 @@ void __cpuinit platform_secondary_init(unsigned int cpu)
@@ -70,8 +70,8 @@ void __cpuinit platform_secondary_init(u
/*
* Synchronise with the boot thread.
*/
@ -103,7 +102,7 @@ index fdec58a..cad6b81 100644
}
static __cpuinit void prepare_cold_cpu(unsigned int cpu)
@@ -107,7 +107,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -108,7 +108,7 @@ int __cpuinit boot_secondary(unsigned in
* set synchronisation state between this boot processor
* and the secondary one
*/
@ -112,7 +111,7 @@ index fdec58a..cad6b81 100644
/*
* The secondary processor is waiting to be released from
@@ -141,7 +141,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -142,7 +142,7 @@ int __cpuinit boot_secondary(unsigned in
* now the secondary core is starting up let it run its
* calibrations, then wait for it to finish
*/
@ -121,20 +120,20 @@ index fdec58a..cad6b81 100644
return pen_release != -1 ? -ENOSYS : 0;
}
diff --git a/arch/arm/mach-omap2/omap-smp.c b/arch/arm/mach-omap2/omap-smp.c
index 4412ddb..490de9c 100644
--- a/arch/arm/mach-omap2/omap-smp.c
+++ b/arch/arm/mach-omap2/omap-smp.c
@@ -29,7 +29,7 @@
Index: linux-3.4/arch/arm/mach-omap2/omap-smp.c
===================================================================
--- linux-3.4.orig/arch/arm/mach-omap2/omap-smp.c
+++ linux-3.4/arch/arm/mach-omap2/omap-smp.c
@@ -34,7 +34,7 @@
/* SCU base address */
static void __iomem *scu_base;
-static DEFINE_SPINLOCK(boot_lock);
+static DEFINE_RAW_SPINLOCK(boot_lock);
void __cpuinit platform_secondary_init(unsigned int cpu)
void __iomem *omap4_get_scu_base(void)
{
@@ -43,8 +43,8 @@ void __cpuinit platform_secondary_init(unsigned int cpu)
@@ -65,8 +65,8 @@ void __cpuinit platform_secondary_init(u
/*
* Synchronise with the boot thread.
*/
@ -145,7 +144,7 @@ index 4412ddb..490de9c 100644
}
int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -53,7 +53,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -77,7 +77,7 @@ int __cpuinit boot_secondary(unsigned in
* Set synchronisation state between this boot processor
* and the secondary one
*/
@ -154,7 +153,7 @@ index 4412ddb..490de9c 100644
/*
* Update the AuxCoreBoot0 with boot state for secondary core.
@@ -70,7 +70,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -117,7 +117,7 @@ int __cpuinit boot_secondary(unsigned in
* Now the secondary core is starting up let it run its
* calibrations, then wait for it to finish
*/
@ -163,53 +162,11 @@ index 4412ddb..490de9c 100644
return 0;
}
diff --git a/arch/arm/mach-tegra/platsmp.c b/arch/arm/mach-tegra/platsmp.c
index 7d2b5d0..571f61a 100644
--- a/arch/arm/mach-tegra/platsmp.c
+++ b/arch/arm/mach-tegra/platsmp.c
@@ -28,7 +28,7 @@
extern void tegra_secondary_startup(void);
-static DEFINE_SPINLOCK(boot_lock);
+static DEFINE_RAW_SPINLOCK(boot_lock);
static void __iomem *scu_base = IO_ADDRESS(TEGRA_ARM_PERIF_BASE);
#define EVP_CPU_RESET_VECTOR \
@@ -50,8 +50,8 @@ void __cpuinit platform_secondary_init(unsigned int cpu)
/*
* Synchronise with the boot thread.
*/
- spin_lock(&boot_lock);
- spin_unlock(&boot_lock);
+ raw_spin_lock(&boot_lock);
+ raw_spin_unlock(&boot_lock);
}
int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -65,7 +65,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
* set synchronisation state between this boot processor
* and the secondary one
*/
- spin_lock(&boot_lock);
+ raw_spin_lock(&boot_lock);
/* set the reset vector to point to the secondary_startup routine */
@@ -101,7 +101,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
* now the secondary core is starting up let it run its
* calibrations, then wait for it to finish
*/
- spin_unlock(&boot_lock);
+ raw_spin_unlock(&boot_lock);
return 0;
}
diff --git a/arch/arm/mach-ux500/platsmp.c b/arch/arm/mach-ux500/platsmp.c
index a19e398..9e92c6c 100644
--- a/arch/arm/mach-ux500/platsmp.c
+++ b/arch/arm/mach-ux500/platsmp.c
@@ -57,7 +57,7 @@ static void __iomem *scu_base_addr(void)
Index: linux-3.4/arch/arm/mach-ux500/platsmp.c
===================================================================
--- linux-3.4.orig/arch/arm/mach-ux500/platsmp.c
+++ linux-3.4/arch/arm/mach-ux500/platsmp.c
@@ -58,7 +58,7 @@ static void __iomem *scu_base_addr(void)
return NULL;
}
@ -218,7 +175,7 @@ index a19e398..9e92c6c 100644
void __cpuinit platform_secondary_init(unsigned int cpu)
{
@@ -77,8 +77,8 @@ void __cpuinit platform_secondary_init(unsigned int cpu)
@@ -78,8 +78,8 @@ void __cpuinit platform_secondary_init(u
/*
* Synchronise with the boot thread.
*/
@ -229,7 +186,7 @@ index a19e398..9e92c6c 100644
}
int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -89,7 +89,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -90,7 +90,7 @@ int __cpuinit boot_secondary(unsigned in
* set synchronisation state between this boot processor
* and the secondary one
*/
@ -238,7 +195,7 @@ index a19e398..9e92c6c 100644
/*
* The secondary processor is waiting to be released from
@@ -110,7 +110,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -111,7 +111,7 @@ int __cpuinit boot_secondary(unsigned in
* now the secondary core is starting up let it run its
* calibrations, then wait for it to finish
*/
@ -247,11 +204,11 @@ index a19e398..9e92c6c 100644
return pen_release != -1 ? -ENOSYS : 0;
}
diff --git a/arch/arm/plat-versatile/platsmp.c b/arch/arm/plat-versatile/platsmp.c
index 92f18d3..287bbb5 100644
--- a/arch/arm/plat-versatile/platsmp.c
+++ b/arch/arm/plat-versatile/platsmp.c
@@ -37,7 +37,7 @@ static void __cpuinit write_pen_release(int val)
Index: linux-3.4/arch/arm/plat-versatile/platsmp.c
===================================================================
--- linux-3.4.orig/arch/arm/plat-versatile/platsmp.c
+++ linux-3.4/arch/arm/plat-versatile/platsmp.c
@@ -38,7 +38,7 @@ static void __cpuinit write_pen_release(
outer_clean_range(__pa(&pen_release), __pa(&pen_release + 1));
}
@ -260,7 +217,7 @@ index 92f18d3..287bbb5 100644
void __cpuinit platform_secondary_init(unsigned int cpu)
{
@@ -57,8 +57,8 @@ void __cpuinit platform_secondary_init(unsigned int cpu)
@@ -58,8 +58,8 @@ void __cpuinit platform_secondary_init(u
/*
* Synchronise with the boot thread.
*/
@ -271,7 +228,7 @@ index 92f18d3..287bbb5 100644
}
int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -69,7 +69,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -70,7 +70,7 @@ int __cpuinit boot_secondary(unsigned in
* Set synchronisation state between this boot processor
* and the secondary one
*/
@ -280,7 +237,7 @@ index 92f18d3..287bbb5 100644
/*
* This is really belt and braces; we hold unintended secondary
@@ -99,7 +99,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct task_struct *idle)
@@ -100,7 +100,7 @@ int __cpuinit boot_secondary(unsigned in
* now the secondary core is starting up let it run its
* calibrations, then wait for it to finish
*/
@ -289,6 +246,3 @@ index 92f18d3..287bbb5 100644
return pen_release != -1 ? -ENOSYS : 0;
}
--
1.7.10

16
debian/patches/features/all/rt/0238-arm-disable-highmem-on-rt.patch.patch → debian/patches/features/all/rt/arm-disable-highmem-on-rt.patch vendored
View File

@ -1,18 +1,17 @@
From 9b87c1e6e04a2419345383133d155c912ec6ae9f Mon Sep 17 00:00:00 2001
Subject: arm-disable-highmem-on-rt.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 18 Jul 2011 17:09:28 +0200
Subject: [PATCH 238/267] arm-disable-highmem-on-rt.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index fc327b8..0e53586 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -1668,7 +1668,7 @@ config HAVE_ARCH_PFN_VALID
Index: linux-3.4/arch/arm/Kconfig
===================================================================
--- linux-3.4.orig/arch/arm/Kconfig
+++ linux-3.4/arch/arm/Kconfig
@@ -1716,7 +1716,7 @@ config HAVE_ARCH_PFN_VALID
config HIGHMEM
bool "High Memory Support"
@ -21,6 +20,3 @@ index fc327b8..0e53586 100644
help
The address space of ARM processors is only 4 Gigabytes large
and it has to accommodate user address space, kernel address
--
1.7.10

23
debian/patches/features/all/rt/arm-enable-interrupts-in-signal-code.patch vendored Normal file
View File

@ -0,0 +1,23 @@
Subject: arm-enable-interrupts-in-signal-code.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 16 Jul 2011 16:27:13 +0200
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/kernel/signal.c | 3 +++
1 file changed, 3 insertions(+)
Index: linux-3.4/arch/arm/kernel/signal.c
===================================================================
--- linux-3.4.orig/arch/arm/kernel/signal.c
+++ linux-3.4/arch/arm/kernel/signal.c
@@ -664,6 +664,9 @@ static void do_signal(struct pt_regs *re
if (!user_mode(regs))
return;
+ local_irq_enable();
+ preempt_check_resched();
+
/*
* If we were from a system call, check for system call restarting...
*/

16
debian/patches/features/all/rt/0027-arm-Mark-pmu-interupt-IRQF_NO_THREAD.patch → debian/patches/features/all/rt/arm-mark-pmu-interupt-no-thread.patch vendored
View File

@ -1,7 +1,6 @@
From 8aa76e7290f80f472b9aebdaf8ae7d627d579359 Mon Sep 17 00:00:00 2001
Subject: arm: Mark pmu interupt IRQF_NO_THREAD
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 16 Mar 2011 14:45:31 +0100
Subject: [PATCH 027/267] arm: Mark pmu interupt IRQF_NO_THREAD
PMU interrupt must not be threaded. Remove IRQF_DISABLED while at it
as we run all handlers with interrupts disabled anyway.
@ -11,11 +10,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arch/arm/kernel/perf_event.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c
index ecebb89..e1286fb 100644
--- a/arch/arm/kernel/perf_event.c
+++ b/arch/arm/kernel/perf_event.c
@@ -435,7 +435,7 @@ armpmu_reserve_hardware(struct arm_pmu *armpmu)
Index: linux-3.4/arch/arm/kernel/perf_event.c
===================================================================
--- linux-3.4.orig/arch/arm/kernel/perf_event.c
+++ linux-3.4/arch/arm/kernel/perf_event.c
@@ -433,7 +433,7 @@ armpmu_reserve_hardware(struct arm_pmu *
}
err = request_irq(irq, handle_irq,
@ -24,6 +23,3 @@ index ecebb89..e1286fb 100644
"arm-pmu", armpmu);
if (err) {
pr_err("unable to request IRQ%d for ARM PMU counters\n",
--
1.7.10

64
debian/patches/features/all/rt/arm-omap-make-wakeupgen_lock-raw.patch vendored Normal file
View File

@ -0,0 +1,64 @@
Subject: arm-omap-make-wakeupgen_lock-raw.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 11 Apr 2012 11:26:38 +0200
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/arm/mach-omap2/omap-wakeupgen.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
Index: linux-3.4/arch/arm/mach-omap2/omap-wakeupgen.c
===================================================================
--- linux-3.4.orig/arch/arm/mach-omap2/omap-wakeupgen.c
+++ linux-3.4/arch/arm/mach-omap2/omap-wakeupgen.c
@@ -43,7 +43,7 @@
static void __iomem *wakeupgen_base;
static void __iomem *sar_base;
-static DEFINE_SPINLOCK(wakeupgen_lock);
+static DEFINE_RAW_SPINLOCK(wakeupgen_lock);
static unsigned int irq_target_cpu[NR_IRQS];
/*
@@ -128,9 +128,9 @@ static void wakeupgen_mask(struct irq_da
{
unsigned long flags;
- spin_lock_irqsave(&wakeupgen_lock, flags);
+ raw_spin_lock_irqsave(&wakeupgen_lock, flags);
_wakeupgen_clear(d->irq, irq_target_cpu[d->irq]);
- spin_unlock_irqrestore(&wakeupgen_lock, flags);
+ raw_spin_unlock_irqrestore(&wakeupgen_lock, flags);
}
/*
@@ -140,9 +140,9 @@ static void wakeupgen_unmask(struct irq_
{
unsigned long flags;
- spin_lock_irqsave(&wakeupgen_lock, flags);
+ raw_spin_lock_irqsave(&wakeupgen_lock, flags);
_wakeupgen_set(d->irq, irq_target_cpu[d->irq]);
- spin_unlock_irqrestore(&wakeupgen_lock, flags);
+ raw_spin_unlock_irqrestore(&wakeupgen_lock, flags);
}
#ifdef CONFIG_HOTPLUG_CPU
@@ -183,7 +183,7 @@ static void wakeupgen_irqmask_all(unsign
{
unsigned long flags;
- spin_lock_irqsave(&wakeupgen_lock, flags);
+ raw_spin_lock_irqsave(&wakeupgen_lock, flags);
if (set) {
_wakeupgen_save_masks(cpu);
_wakeupgen_set_all(cpu, WKG_MASK_ALL);
@@ -191,7 +191,7 @@ static void wakeupgen_irqmask_all(unsign
_wakeupgen_set_all(cpu, WKG_UNMASK_ALL);
_wakeupgen_restore_masks(cpu);
}
- spin_unlock_irqrestore(&wakeupgen_lock, flags);
+ raw_spin_unlock_irqrestore(&wakeupgen_lock, flags);
}
#endif

30
debian/patches/features/all/rt/0080-ata-Do-not-disable-interrupts-in-ide-code-for-preemp.patch → debian/patches/features/all/rt/ata-disable-interrupts-if-non-rt.patch vendored
View File

@ -1,23 +1,22 @@
From 8568db0e556b5159c9c7d85ce78f66d1df5d50ee Mon Sep 17 00:00:00 2001
From: Steven Rostedt <srostedt@redhat.com>
Date: Fri, 3 Jul 2009 08:44:29 -0500
Subject: [PATCH 080/267] ata: Do not disable interrupts in ide code for
preempt-rt
Subject: ata: Do not disable interrupts in ide code for preempt-rt
Use the local_irq_*_nort variants.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/ata/libata-sff.c | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index 4cadfa2..100f85c 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -678,9 +678,9 @@ unsigned int ata_sff_data_xfer_noirq(struct ata_device *dev, unsigned char *buf,
Index: linux-3.4/drivers/ata/libata-sff.c
===================================================================
--- linux-3.4.orig/drivers/ata/libata-sff.c
+++ linux-3.4/drivers/ata/libata-sff.c
@@ -678,9 +678,9 @@ unsigned int ata_sff_data_xfer_noirq(str
unsigned long flags;
unsigned int consumed;
@ -29,19 +28,19 @@ index 4cadfa2..100f85c 100644
return consumed;
}
@@ -719,7 +719,7 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
@@ -719,7 +719,7 @@ static void ata_pio_sector(struct ata_qu
unsigned long flags;
/* FIXME: use a bounce buffer */
- local_irq_save(flags);
+ local_irq_save_nort(flags);
buf = kmap_atomic(page, KM_IRQ0);
buf = kmap_atomic(page);
/* do the actual data transfer */
@@ -727,7 +727,7 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
@@ -727,7 +727,7 @@ static void ata_pio_sector(struct ata_qu
do_write);
kunmap_atomic(buf, KM_IRQ0);
kunmap_atomic(buf);
- local_irq_restore(flags);
+ local_irq_restore_nort(flags);
} else {
@ -53,18 +52,15 @@ index 4cadfa2..100f85c 100644
/* FIXME: use bounce buffer */
- local_irq_save(flags);
+ local_irq_save_nort(flags);
buf = kmap_atomic(page, KM_IRQ0);
buf = kmap_atomic(page);
/* do the actual data transfer */
@@ -872,7 +872,7 @@ next_sg:
count, rw);
kunmap_atomic(buf, KM_IRQ0);
kunmap_atomic(buf);
- local_irq_restore(flags);
+ local_irq_restore_nort(flags);
} else {
buf = page_address(page);
consumed = ap->ops->sff_data_xfer(dev, buf + offset,
--
1.7.10

28
debian/patches/features/all/rt/0005-block-Shorten-interrupt-disabled-regions.patch → debian/patches/features/all/rt/block-shorten-interrupt-disabled-regions.patch vendored
View File

@ -1,7 +1,6 @@
From e2c97e4d5da80e66340ea752cb33bf8802feaa53 Mon Sep 17 00:00:00 2001
Subject: block: Shorten interrupt disabled regions
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 22 Jun 2011 19:47:02 +0200
Subject: [PATCH 005/267] block: Shorten interrupt disabled regions
Moving the blk_sched_flush_plug() call out of the interrupt/preempt
disabled region in the scheduler allows us to replace
@ -47,11 +46,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
block/blk-core.c | 20 ++++++++------------
1 file changed, 8 insertions(+), 12 deletions(-)
diff --git a/block/blk-core.c b/block/blk-core.c
index 15de223..7366ad4 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -300,7 +300,11 @@ void __blk_run_queue(struct request_queue *q)
Index: linux-3.4/block/blk-core.c
===================================================================
--- linux-3.4.orig/block/blk-core.c
+++ linux-3.4/block/blk-core.c
@@ -302,7 +302,11 @@ void __blk_run_queue(struct request_queu
{
if (unlikely(blk_queue_stopped(q)))
return;
@ -64,7 +63,7 @@ index 15de223..7366ad4 100644
q->request_fn(q);
}
EXPORT_SYMBOL(__blk_run_queue);
@@ -2745,11 +2749,11 @@ static void queue_unplugged(struct request_queue *q, unsigned int depth,
@@ -2779,11 +2783,11 @@ static void queue_unplugged(struct reque
* this lock).
*/
if (from_schedule) {
@ -78,7 +77,7 @@ index 15de223..7366ad4 100644
}
}
@@ -2775,7 +2779,6 @@ static void flush_plug_callbacks(struct blk_plug *plug)
@@ -2809,7 +2813,6 @@ static void flush_plug_callbacks(struct
void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
{
struct request_queue *q;
@ -86,7 +85,7 @@ index 15de223..7366ad4 100644
struct request *rq;
LIST_HEAD(list);
unsigned int depth;
@@ -2796,11 +2799,6 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
@@ -2830,11 +2833,6 @@ void blk_flush_plug_list(struct blk_plug
q = NULL;
depth = 0;
@ -98,16 +97,16 @@ index 15de223..7366ad4 100644
while (!list_empty(&list)) {
rq = list_entry_rq(list.next);
list_del_init(&rq->queuelist);
@@ -2813,7 +2811,7 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
@@ -2847,7 +2845,7 @@ void blk_flush_plug_list(struct blk_plug
queue_unplugged(q, depth, from_schedule);
q = rq->q;
depth = 0;
- spin_lock(q->queue_lock);
+ spin_lock_irq(q->queue_lock);
}
/*
* rq is already accounted, so use raw insert
@@ -2831,8 +2829,6 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
@@ -2874,8 +2872,6 @@ void blk_flush_plug_list(struct blk_plug
*/
if (q)
queue_unplugged(q, depth, from_schedule);
@ -116,6 +115,3 @@ index 15de223..7366ad4 100644
}
void blk_finish_plug(struct blk_plug *plug)
--
1.7.10

17
debian/patches/features/all/rt/0077-bug-BUG_ON-WARN_ON-variants-dependend-on-RT-RT.patch → debian/patches/features/all/rt/bug-rt-dependend-variants.patch vendored
View File

@ -1,18 +1,18 @@
From 7258f0c5442a0dde07c06e12f431134e4b9bff47 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:58 -0500
Subject: [PATCH 077/267] bug: BUG_ON/WARN_ON variants dependend on RT/!RT
Subject: bug: BUG_ON/WARN_ON variants dependend on RT/!RT
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/asm-generic/bug.h | 18 ++++++++++++++++++
1 file changed, 18 insertions(+)
diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index 84458b0..97c1eaf 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
Index: linux-3.4/include/asm-generic/bug.h
===================================================================
--- linux-3.4.orig/include/asm-generic/bug.h
+++ linux-3.4/include/asm-generic/bug.h
@@ -3,6 +3,10 @@
#include <linux/compiler.h>
@ -24,7 +24,7 @@ index 84458b0..97c1eaf 100644
#ifdef CONFIG_BUG
#ifdef CONFIG_GENERIC_BUG
@@ -202,4 +206,18 @@ extern void warn_slowpath_null(const char *file, const int line);
@@ -202,4 +206,18 @@ extern void warn_slowpath_null(const cha
# define WARN_ON_SMP(x) ({0;})
#endif
@ -43,6 +43,3 @@ index 84458b0..97c1eaf 100644
+#endif
+
#endif
--
1.7.10

54
debian/patches/features/all/rt/0039-clocksource-TCLIB-Allow-higher-clock-rates-for-clock.patch → debian/patches/features/all/rt/clocksource-tclib-allow-higher-clockrates.patch vendored
View File

@ -1,25 +1,24 @@
From b4217e96b0c71c84b8edbca7b99813b5f0f72033 Mon Sep 17 00:00:00 2001
From: Benedikt Spranger <b.spranger@linutronix.de>
Date: Mon, 8 Mar 2010 18:57:04 +0100
Subject: [PATCH 039/267] clocksource: TCLIB: Allow higher clock rates for
clock events
Subject: clocksource: TCLIB: Allow higher clock rates for clock events
As default the TCLIB uses the 32KiHz base clock rate for clock events.
Add a compile time selection to allow higher clock resulution.
Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/clocksource/tcb_clksrc.c | 44 ++++++++++++++++++++++----------------
drivers/misc/Kconfig | 11 ++++++++--
drivers/clocksource/tcb_clksrc.c | 44 +++++++++++++++++++++++----------------
drivers/misc/Kconfig | 11 +++++++--
2 files changed, 35 insertions(+), 20 deletions(-)
diff --git a/drivers/clocksource/tcb_clksrc.c b/drivers/clocksource/tcb_clksrc.c
index 79c47e8..8976b3d 100644
--- a/drivers/clocksource/tcb_clksrc.c
+++ b/drivers/clocksource/tcb_clksrc.c
@@ -21,8 +21,7 @@
* resolution better than 200 nsec).
Index: linux-3.4/drivers/clocksource/tcb_clksrc.c
===================================================================
--- linux-3.4.orig/drivers/clocksource/tcb_clksrc.c
+++ linux-3.4/drivers/clocksource/tcb_clksrc.c
@@ -23,8 +23,7 @@
* this 32 bit free-running counter. the second channel is not used.
*
* - The third channel may be used to provide a 16-bit clockevent
- * source, used in either periodic or oneshot mode. This runs
@ -28,7 +27,7 @@ index 79c47e8..8976b3d 100644
*
* A boot clocksource and clockevent source are also currently needed,
* unless the relevant platforms (ARM/AT91, AVR32/AT32) are changed so
@@ -68,6 +67,7 @@ static struct clocksource clksrc = {
@@ -74,6 +73,7 @@ static struct clocksource clksrc = {
struct tc_clkevt_device {
struct clock_event_device clkevt;
struct clk *clk;
@ -36,7 +35,7 @@ index 79c47e8..8976b3d 100644
void __iomem *regs;
};
@@ -76,13 +76,6 @@ static struct tc_clkevt_device *to_tc_clkevt(struct clock_event_device *clkevt)
@@ -82,13 +82,6 @@ static struct tc_clkevt_device *to_tc_cl
return container_of(clkevt, struct tc_clkevt_device, clkevt);
}
@ -50,7 +49,7 @@ index 79c47e8..8976b3d 100644
static u32 timer_clock;
static void tc_mode(enum clock_event_mode m, struct clock_event_device *d)
@@ -105,11 +98,12 @@ static void tc_mode(enum clock_event_mode m, struct clock_event_device *d)
@@ -111,11 +104,12 @@ static void tc_mode(enum clock_event_mod
case CLOCK_EVT_MODE_PERIODIC:
clk_enable(tcd->clk);
@ -65,7 +64,7 @@ index 79c47e8..8976b3d 100644
/* Enable clock and interrupts on RC compare */
__raw_writel(ATMEL_TC_CPCS, regs + ATMEL_TC_REG(2, IER));
@@ -122,7 +116,7 @@ static void tc_mode(enum clock_event_mode m, struct clock_event_device *d)
@@ -128,7 +122,7 @@ static void tc_mode(enum clock_event_mod
case CLOCK_EVT_MODE_ONESHOT:
clk_enable(tcd->clk);
@ -74,7 +73,7 @@ index 79c47e8..8976b3d 100644
__raw_writel(timer_clock | ATMEL_TC_CPCSTOP
| ATMEL_TC_WAVE | ATMEL_TC_WAVESEL_UP_AUTO,
regs + ATMEL_TC_REG(2, CMR));
@@ -152,8 +146,12 @@ static struct tc_clkevt_device clkevt = {
@@ -158,8 +152,12 @@ static struct tc_clkevt_device clkevt =
.features = CLOCK_EVT_FEAT_PERIODIC
| CLOCK_EVT_FEAT_ONESHOT,
.shift = 32,
@ -87,7 +86,7 @@ index 79c47e8..8976b3d 100644
.set_next_event = tc_next_event,
.set_mode = tc_mode,
},
@@ -179,8 +177,9 @@ static struct irqaction tc_irqaction = {
@@ -185,8 +183,9 @@ static struct irqaction tc_irqaction = {
.handler = ch2_irq,
};
@ -98,7 +97,7 @@ index 79c47e8..8976b3d 100644
struct clk *t2_clk = tc->clk[2];
int irq = tc->irq[2];
@@ -188,11 +187,17 @@ static void __init setup_clkevents(struct atmel_tc *tc, int clk32k_divisor_idx)
@@ -194,11 +193,17 @@ static void __init setup_clkevents(struc
clkevt.clk = t2_clk;
tc_irqaction.dev_id = &clkevt;
@ -120,8 +119,8 @@ index 79c47e8..8976b3d 100644
clkevt.clkevt.min_delta_ns = clockevent_delta2ns(1, &clkevt.clkevt) + 1;
clkevt.clkevt.cpumask = cpumask_of(0);
@@ -295,8 +300,11 @@ static int __init tcb_clksrc_init(void)
clocksource_register(&clksrc);
@@ -327,8 +332,11 @@ static int __init tcb_clksrc_init(void)
clocksource_register_hz(&clksrc, divided_rate);
/* channel 2: periodic and oneshot timer support */
+#ifdef CONFIG_ATMEL_TCB_CLKSRC_USE_SLOW_CLOCK
@ -133,11 +132,11 @@ index 79c47e8..8976b3d 100644
return 0;
}
arch_initcall(tcb_clksrc_init);
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 5664696..f3031a4 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -97,8 +97,7 @@ config ATMEL_TCB_CLKSRC
Index: linux-3.4/drivers/misc/Kconfig
===================================================================
--- linux-3.4.orig/drivers/misc/Kconfig
+++ linux-3.4/drivers/misc/Kconfig
@@ -87,8 +87,7 @@ config ATMEL_TCB_CLKSRC
are combined to make a single 32-bit timer.
When GENERIC_CLOCKEVENTS is defined, the third timer channel
@ -147,7 +146,7 @@ index 5664696..f3031a4 100644
config ATMEL_TCB_CLKSRC_BLOCK
int
@@ -112,6 +111,14 @@ config ATMEL_TCB_CLKSRC_BLOCK
@@ -102,6 +101,14 @@ config ATMEL_TCB_CLKSRC_BLOCK
TC can be used for other purposes, such as PWM generation and
interval timing.
@ -162,6 +161,3 @@ index 5664696..f3031a4 100644
config IBM_ASM
tristate "Device driver for IBM RSA service processor"
depends on X86 && PCI && INPUT && EXPERIMENTAL
--
1.7.10

16
debian/patches/features/all/rt/0151-cond-resched-lock-rt-tweak.patch.patch → debian/patches/features/all/rt/cond-resched-lock-rt-tweak.patch vendored
View File

@ -1,18 +1,17 @@
From 5136313c61afded30a8da141ae60d4bca90b0cf2 Mon Sep 17 00:00:00 2001
Subject: cond-resched-lock-rt-tweak.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 22:51:33 +0200
Subject: [PATCH 151/267] cond-resched-lock-rt-tweak.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/sched.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 99e7f8b..175aaee 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2599,7 +2599,7 @@ extern int _cond_resched(void);
Index: linux-3.4/include/linux/sched.h
===================================================================
--- linux-3.4.orig/include/linux/sched.h
+++ linux-3.4/include/linux/sched.h
@@ -2689,7 +2689,7 @@ extern int _cond_resched(void);
extern int __cond_resched_lock(spinlock_t *lock);
@ -21,6 +20,3 @@ index 99e7f8b..175aaee 100644
#define PREEMPT_LOCK_OFFSET PREEMPT_OFFSET
#else
#define PREEMPT_LOCK_OFFSET 0
--
1.7.10

30
debian/patches/features/all/rt/0149-cond-resched-softirq-fix.patch.patch → debian/patches/features/all/rt/cond-resched-softirq-rt.patch vendored
View File

@ -1,19 +1,18 @@
From 1a9f6b8b37a41e2b59c1989b461bf5b1fbecb50e Mon Sep 17 00:00:00 2001
Subject: cond-resched-softirq-fix.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 14 Jul 2011 09:56:44 +0200
Subject: [PATCH 149/267] cond-resched-softirq-fix.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/sched.h | 4 ++++
kernel/sched.c | 2 ++
kernel/sched/core.c | 2 ++
2 files changed, 6 insertions(+)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 6c20349..38d78dc 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2602,12 +2602,16 @@ extern int __cond_resched_lock(spinlock_t *lock);
Index: linux-3.4/include/linux/sched.h
===================================================================
--- linux-3.4.orig/include/linux/sched.h
+++ linux-3.4/include/linux/sched.h
@@ -2700,12 +2700,16 @@ extern int __cond_resched_lock(spinlock_
__cond_resched_lock(lock); \
})
@ -30,11 +29,11 @@ index 6c20349..38d78dc 100644
/*
* Does a critical section need to be broken due to another
diff --git a/kernel/sched.c b/kernel/sched.c
index 42d0003..af3883b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5826,6 +5826,7 @@ int __cond_resched_lock(spinlock_t *lock)
Index: linux-3.4/kernel/sched/core.c
===================================================================
--- linux-3.4.orig/kernel/sched/core.c
+++ linux-3.4/kernel/sched/core.c
@@ -4605,6 +4605,7 @@ int __cond_resched_lock(spinlock_t *lock
}
EXPORT_SYMBOL(__cond_resched_lock);
@ -42,7 +41,7 @@ index 42d0003..af3883b 100644
int __sched __cond_resched_softirq(void)
{
BUG_ON(!in_softirq());
@@ -5839,6 +5840,7 @@ int __sched __cond_resched_softirq(void)
@@ -4618,6 +4619,7 @@ int __sched __cond_resched_softirq(void)
return 0;
}
EXPORT_SYMBOL(__cond_resched_softirq);
@ -50,6 +49,3 @@ index 42d0003..af3883b 100644
/**
* yield - yield the current processor to other threads.
--
1.7.10

24
debian/patches/features/all/rt/0234-console-make-rt-friendly.patch.patch → debian/patches/features/all/rt/console-make-rt-friendly.patch vendored
View File

@ -1,18 +1,17 @@
From 7345e11db5653c75894b0a3d3cf2f0f380016b79 Mon Sep 17 00:00:00 2001
Subject: console-make-rt-friendly.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 22:43:07 +0200
Subject: [PATCH 234/267] console-make-rt-friendly.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/printk.c | 26 +++++++++++++++++++++++---
1 file changed, 23 insertions(+), 3 deletions(-)
diff --git a/kernel/printk.c b/kernel/printk.c
index 2b95bc0..fad23c8 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -504,6 +504,7 @@ static void __call_console_drivers(unsigned start, unsigned end)
Index: linux-3.4/kernel/printk.c
===================================================================
--- linux-3.4.orig/kernel/printk.c
+++ linux-3.4/kernel/printk.c
@@ -507,6 +507,7 @@ static void __call_console_drivers(unsig
{
struct console *con;
@ -20,7 +19,7 @@ index 2b95bc0..fad23c8 100644
for_each_console(con) {
if (exclusive_console && con != exclusive_console)
continue;
@@ -512,6 +513,7 @@ static void __call_console_drivers(unsigned start, unsigned end)
@@ -515,6 +516,7 @@ static void __call_console_drivers(unsig
(con->flags & CON_ANYTIME)))
con->write(con, &LOG_BUF(start), end - start);
}
@ -28,7 +27,7 @@ index 2b95bc0..fad23c8 100644
}
#ifdef CONFIG_EARLY_PRINTK
@@ -827,12 +829,18 @@ static inline int can_use_console(unsigned int cpu)
@@ -839,12 +841,18 @@ static inline int can_use_console(unsign
* interrupts disabled. It should return with 'lockbuf_lock'
* released but interrupts still disabled.
*/
@ -49,7 +48,7 @@ index 2b95bc0..fad23c8 100644
retval = 1;
/*
@@ -1010,8 +1018,15 @@ asmlinkage int vprintk(const char *fmt, va_list args)
@@ -1021,8 +1029,15 @@ asmlinkage int vprintk(const char *fmt,
* will release 'logbuf_lock' regardless of whether it
* actually gets the semaphore or not.
*/
@ -66,7 +65,7 @@ index 2b95bc0..fad23c8 100644
lockdep_on();
out_restore_irqs:
@@ -1321,11 +1336,16 @@ again:
@@ -1345,11 +1360,16 @@ again:
_con_start = con_start;
_log_end = log_end;
con_start = log_end; /* Flush */
@ -83,6 +82,3 @@ index 2b95bc0..fad23c8 100644
}
console_locked = 0;
--
1.7.10

16
debian/patches/features/all/rt/0259-cpu-Make-hotplug.lock-a-sleeping-spinlock-on-RT.patch → debian/patches/features/all/rt/cpu-rt-make-hotplug-lock-a-sleeping-spinlock-on-rt.patch vendored
View File

@ -1,7 +1,6 @@
From 7414f6f107d40967dc5fe7339f11c70ce57e8b74 Mon Sep 17 00:00:00 2001
Subject: cpu: Make hotplug.lock a "sleeping" spinlock on RT
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 2 Mar 2012 10:36:57 -0500
Subject: [PATCH 259/267] cpu: Make hotplug.lock a "sleeping" spinlock on RT
Date: Fri, 02 Mar 2012 10:36:57 -0500
Tasks can block on hotplug.lock in pin_current_cpu(), but their state
might be != RUNNING. So the mutex wakeup will set the state
@ -23,10 +22,10 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kernel/cpu.c | 35 ++++++++++++++++++++++++++---------
1 file changed, 26 insertions(+), 9 deletions(-)
diff --git a/kernel/cpu.c b/kernel/cpu.c
index fa40834..66dfb74 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
Index: linux-3.4/kernel/cpu.c
===================================================================
--- linux-3.4.orig/kernel/cpu.c
+++ linux-3.4/kernel/cpu.c
@@ -46,7 +46,12 @@ static int cpu_hotplug_disabled;
static struct {
@ -122,6 +121,3 @@ index fa40834..66dfb74 100644
}
#else /* #if CONFIG_HOTPLUG_CPU */
--
1.7.10

16
debian/patches/features/all/rt/0105-cpu-rt-variants.patch.patch → debian/patches/features/all/rt/cpu-rt-variants.patch vendored
View File

@ -1,18 +1,17 @@
From 4999ba9453b30762e55beaf253c8999c58266fc3 Mon Sep 17 00:00:00 2001
Subject: cpu-rt-variants.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 17 Jun 2011 15:42:38 +0200
Subject: [PATCH 105/267] cpu-rt-variants.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/linux/smp.h | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 8cc38d3..e6c58d8 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -173,6 +173,14 @@ smp_call_function_any(const struct cpumask *mask, smp_call_func_t func,
Index: linux-3.4/include/linux/smp.h
===================================================================
--- linux-3.4.orig/include/linux/smp.h
+++ linux-3.4/include/linux/smp.h
@@ -219,6 +219,14 @@ smp_call_function_any(const struct cpuma
#define get_cpu() ({ preempt_disable(); smp_processor_id(); })
#define put_cpu() preempt_enable()
@ -27,6 +26,3 @@ index 8cc38d3..e6c58d8 100644
/*
* Callback to arch code if there's nosmp or maxcpus=0 on the
* boot command line:
--
1.7.10

26
debian/patches/features/all/rt/0252-cpumask-Disable-CONFIG_CPUMASK_OFFSTACK-for-RT.patch → debian/patches/features/all/rt/cpumask-disable-offstack-on-rt.patch vendored
View File

@ -1,7 +1,6 @@
From 113ec250cd47567fa333e584f5a210780df47ac7 Mon Sep 17 00:00:00 2001
Subject: cpumask: Disable CONFIG_CPUMASK_OFFSTACK for RT
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 14 Dec 2011 01:03:49 +0100
Subject: [PATCH 252/267] cpumask: Disable CONFIG_CPUMASK_OFFSTACK for RT
We can't deal with the cpumask allocations which happen in atomic
context (see arch/x86/kernel/apic/io_apic.c) on RT right now.
@ -12,11 +11,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
lib/Kconfig | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e084a73..c42146e 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -730,7 +730,7 @@ config IOMMU_HELPER
Index: linux-3.4/arch/x86/Kconfig
===================================================================
--- linux-3.4.orig/arch/x86/Kconfig
+++ linux-3.4/arch/x86/Kconfig
@@ -750,7 +750,7 @@ config IOMMU_HELPER
config MAXSMP
bool "Enable Maximum number of SMP Processors and NUMA Nodes"
depends on X86_64 && SMP && DEBUG_KERNEL && EXPERIMENTAL
@ -25,11 +24,11 @@ index e084a73..c42146e 100644
---help---
Enable maximum number of CPUS and NUMA Nodes for this architecture.
If unsure, say N.
diff --git a/lib/Kconfig b/lib/Kconfig
index 32f3e5a..63d81e8 100644
--- a/lib/Kconfig
+++ b/lib/Kconfig
@@ -231,6 +231,7 @@ config CHECK_SIGNATURE
Index: linux-3.4/lib/Kconfig
===================================================================
--- linux-3.4.orig/lib/Kconfig
+++ linux-3.4/lib/Kconfig
@@ -303,6 +303,7 @@ config CHECK_SIGNATURE
config CPUMASK_OFFSTACK
bool "Force CPU masks off stack" if DEBUG_PER_CPU_MAPS
@ -37,6 +36,3 @@ index 32f3e5a..63d81e8 100644
help
Use dynamic allocation for cpumask_var_t, instead of putting
them on the stack. This is a bit more expensive, but avoids
--
1.7.10

18
debian/patches/features/all/rt/0230-debugobjects-rt.patch.patch → debian/patches/features/all/rt/debugobjects-rt.patch vendored
View File

@ -1,18 +1,17 @@
From 0fbe4d83b24b5cb2db0da8b4ddd3fa545013bd93 Mon Sep 17 00:00:00 2001
Subject: debugobjects-rt.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 21:41:35 +0200
Subject: [PATCH 230/267] debugobjects-rt.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
lib/debugobjects.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/lib/debugobjects.c b/lib/debugobjects.c
index a78b7c6..9b622c9 100644
--- a/lib/debugobjects.c
+++ b/lib/debugobjects.c
@@ -306,7 +306,10 @@ __debug_object_init(void *addr, struct debug_obj_descr *descr, int onstack)
Index: linux-3.4/lib/debugobjects.c
===================================================================
--- linux-3.4.orig/lib/debugobjects.c
+++ linux-3.4/lib/debugobjects.c
@@ -310,7 +310,10 @@ __debug_object_init(void *addr, struct d
struct debug_obj *obj;
unsigned long flags;
@ -24,7 +23,7 @@ index a78b7c6..9b622c9 100644
db = get_bucket((unsigned long) addr);
@@ -1015,9 +1018,9 @@ static int __init debug_objects_replace_static_objects(void)
@@ -1053,9 +1056,9 @@ static int __init debug_objects_replace_
}
}
@ -35,6 +34,3 @@ index a78b7c6..9b622c9 100644
return 0;
free:
hlist_for_each_entry_safe(obj, node, tmp, &objects, node) {
--
1.7.10

16
debian/patches/features/all/rt/0251-dm-Make-rt-aware.patch → debian/patches/features/all/rt/dm-make-rt-aware.patch vendored
View File

@ -1,7 +1,6 @@
From 7c5e9c5bb3dbe26f8ecc18020a36ed55e99c6a25 Mon Sep 17 00:00:00 2001
Subject: dm: Make rt aware
From: Thomas Gleixner <tglx@linutronix.de>
Date: Mon, 14 Nov 2011 23:06:09 +0100
Subject: [PATCH 251/267] dm: Make rt aware
Use the BUG_ON_NORT variant for the irq_disabled() checks. RT has
interrupts legitimately enabled here as we cant deadlock against the
@ -14,11 +13,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
drivers/md/dm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 4720f68..b1eff42 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1648,14 +1648,14 @@ static void dm_request_fn(struct request_queue *q)
Index: linux-3.4/drivers/md/dm.c
===================================================================
--- linux-3.4.orig/drivers/md/dm.c
+++ linux-3.4/drivers/md/dm.c
@@ -1648,14 +1648,14 @@ static void dm_request_fn(struct request
if (map_request(ti, clone, md))
goto requeued;
@ -35,6 +34,3 @@ index 4720f68..b1eff42 100644
spin_lock(q->queue_lock);
delay_and_out:
--
1.7.10

17
debian/patches/features/all/rt/0041-drivers-net-Use-disable_irq_nosync-in-8139too.patch → debian/patches/features/all/rt/drivers-net-8139-disable-irq-nosync.patch vendored
View File

@ -1,22 +1,22 @@
From 4d7e0a71990ab73749dfde4f4cd201cf1bdb0f45 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:24 -0500
Subject: [PATCH 041/267] drivers/net: Use disable_irq_nosync() in 8139too
Subject: drivers/net: Use disable_irq_nosync() in 8139too
Use disable_irq_nosync() instead of disable_irq() as this might be
called in atomic context with netpoll.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/net/ethernet/realtek/8139too.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c
index 4d6b254..004c054 100644
--- a/drivers/net/ethernet/realtek/8139too.c
+++ b/drivers/net/ethernet/realtek/8139too.c
@@ -2174,7 +2174,7 @@ static irqreturn_t rtl8139_interrupt (int irq, void *dev_instance)
Index: linux-3.4/drivers/net/ethernet/realtek/8139too.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/realtek/8139too.c
+++ linux-3.4/drivers/net/ethernet/realtek/8139too.c
@@ -2240,7 +2240,7 @@ static irqreturn_t rtl8139_interrupt (in
*/
static void rtl8139_poll_controller(struct net_device *dev)
{
@ -25,6 +25,3 @@ index 4d6b254..004c054 100644
rtl8139_interrupt(dev->irq, dev);
enable_irq(dev->irq);
}
--
1.7.10

22
debian/patches/features/all/rt/0043-drivers-net-at91_ether-Make-mdio-protection-rt-safe.patch → debian/patches/features/all/rt/drivers-net-at91-make-mdio-protection-rt-safe.patch vendored
View File

@ -1,8 +1,6 @@
From 2989477b9c1c1f9aa82a8db158ab3a03a4b7b2f5 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 17 Nov 2009 12:02:43 +0100
Subject: [PATCH 043/267] drivers: net: at91_ether: Make mdio protection -rt
safe
Subject: drivers: net: at91_ether: Make mdio protection -rt safe
Neither the phy interrupt nor the timer callback which updates the
link status in absense of a phy interrupt are taking lp->lock which
@ -15,15 +13,16 @@ Fix this by adding proper locking to at91ether_phy_interrupt() and
at91_check_ether() which serializes the access on -rt.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/net/ethernet/cadence/at91_ether.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/drivers/net/ethernet/cadence/at91_ether.c b/drivers/net/ethernet/cadence/at91_ether.c
index 56624d3..ad4dbea 100644
--- a/drivers/net/ethernet/cadence/at91_ether.c
+++ b/drivers/net/ethernet/cadence/at91_ether.c
@@ -200,7 +200,9 @@ static irqreturn_t at91ether_phy_interrupt(int irq, void *dev_id)
Index: linux-3.4/drivers/net/ethernet/cadence/at91_ether.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/cadence/at91_ether.c
+++ linux-3.4/drivers/net/ethernet/cadence/at91_ether.c
@@ -201,7 +201,9 @@ static irqreturn_t at91ether_phy_interru
struct net_device *dev = (struct net_device *) dev_id;
struct at91_private *lp = netdev_priv(dev);
unsigned int phy;
@ -33,7 +32,7 @@ index 56624d3..ad4dbea 100644
/*
* This hander is triggered on both edges, but the PHY chips expect
* level-triggering. We therefore have to check if the PHY actually has
@@ -242,6 +244,7 @@ static irqreturn_t at91ether_phy_interrupt(int irq, void *dev_id)
@@ -243,6 +245,7 @@ static irqreturn_t at91ether_phy_interru
done:
disable_mdi();
@ -41,7 +40,7 @@ index 56624d3..ad4dbea 100644
return IRQ_HANDLED;
}
@@ -398,9 +401,11 @@ static void at91ether_check_link(unsigned long dev_id)
@@ -399,9 +402,11 @@ static void at91ether_check_link(unsigne
struct net_device *dev = (struct net_device *) dev_id;
struct at91_private *lp = netdev_priv(dev);
@ -53,6 +52,3 @@ index 56624d3..ad4dbea 100644
mod_timer(&lp->check_timer, jiffies + LINK_POLL_INTERVAL);
}
--
1.7.10

18
debian/patches/features/all/rt/0042-drivers-net-ehea-Make-rx-irq-handler-non-threaded-IR.patch → debian/patches/features/all/rt/drivers-net-ehea-mark-rx-irq-no-thread.patch vendored
View File

@ -1,8 +1,6 @@
From 1c36d3306722a2042bd4dc26baf20393c42c5b22 Mon Sep 17 00:00:00 2001
From: Darren Hart <dvhltc@us.ibm.com>
Date: Tue, 18 May 2010 14:33:07 -0700
Subject: [PATCH 042/267] drivers: net: ehea: Make rx irq handler non-threaded
(IRQF_NO_THREAD)
Subject: drivers: net: ehea: Make rx irq handler non-threaded (IRQF_NO_THREAD)
The underlying hardware is edge triggered but presented by XICS as level
triggered. The edge triggered interrupts are not reissued after masking. This
@ -35,15 +33,16 @@ Cc: Michael Ellerman <ellerman@au1.ibm.com>
Cc: Doug Maxey <doug.maxey@us.ibm.com>
LKML-Reference: <4BF30793.5070300@us.ibm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/net/ethernet/ibm/ehea/ehea_main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/ibm/ehea/ehea_main.c b/drivers/net/ethernet/ibm/ehea/ehea_main.c
index bfeccbf..fddfaf1 100644
--- a/drivers/net/ethernet/ibm/ehea/ehea_main.c
+++ b/drivers/net/ethernet/ibm/ehea/ehea_main.c
@@ -1304,7 +1304,7 @@ static int ehea_reg_interrupts(struct net_device *dev)
Index: linux-3.4/drivers/net/ethernet/ibm/ehea/ehea_main.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/ibm/ehea/ehea_main.c
+++ linux-3.4/drivers/net/ethernet/ibm/ehea/ehea_main.c
@@ -1306,7 +1306,7 @@ static int ehea_reg_interrupts(struct ne
"%s-queue%d", dev->name, i);
ret = ibmebus_request_irq(pr->eq->attr.ist1,
ehea_recv_irq_handler,
@ -52,6 +51,3 @@ index bfeccbf..fddfaf1 100644
pr);
if (ret) {
netdev_err(dev, "failed registering irq for ehea_queue port_res_nr:%d, ist=%X\n",
--
1.7.10

77
debian/patches/features/all/rt/0099-drivers-net-fix-livelock-issues.patch → debian/patches/features/all/rt/drivers-net-fix-livelock-issues.patch vendored
View File

@ -1,7 +1,6 @@
From 91639026470adfb7ab98a86b62dfefedb485e3a4 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 20 Jun 2009 11:36:54 +0200
Subject: [PATCH 099/267] drivers/net: fix livelock issues
Subject: drivers/net: fix livelock issues
Preempt-RT runs into a live lock issue with the NETDEV_TX_LOCKED micro
optimization. The reason is that the softirq thread is rescheduling
@ -11,6 +10,7 @@ monoplize the CPU and livelock on UP systems.
Remove it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/net/ethernet/atheros/atl1c/atl1c_main.c | 6 +-----
drivers/net/ethernet/atheros/atl1e/atl1e_main.c | 3 +--
@ -21,11 +21,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
drivers/net/rionet.c | 6 +-----
7 files changed, 10 insertions(+), 31 deletions(-)
diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
index eccdcff..ee8d8a2 100644
--- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -2236,11 +2236,7 @@ static netdev_tx_t atl1c_xmit_frame(struct sk_buff *skb,
Index: linux-3.4/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
+++ linux-3.4/drivers/net/ethernet/atheros/atl1c/atl1c_main.c
@@ -2240,11 +2240,7 @@ static netdev_tx_t atl1c_xmit_frame(stru
}
tpd_req = atl1c_cal_tpd_req(skb);
@ -38,11 +38,11 @@ index eccdcff..ee8d8a2 100644
if (atl1c_tpd_avail(adapter, type) < tpd_req) {
/* no enough descriptor, just stop queue */
diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
index 95483bc..eaf84e9 100644
--- a/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
+++ b/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
@@ -1819,8 +1819,7 @@ static netdev_tx_t atl1e_xmit_frame(struct sk_buff *skb,
Index: linux-3.4/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
+++ linux-3.4/drivers/net/ethernet/atheros/atl1e/atl1e_main.c
@@ -1822,8 +1822,7 @@ static netdev_tx_t atl1e_xmit_frame(stru
return NETDEV_TX_OK;
}
tpd_req = atl1e_cal_tdp_req(skb);
@ -52,11 +52,11 @@ index 95483bc..eaf84e9 100644
if (atl1e_tpd_avail(adapter) < tpd_req) {
/* no enough descriptor, just stop queue */
diff --git a/drivers/net/ethernet/chelsio/cxgb/sge.c b/drivers/net/ethernet/chelsio/cxgb/sge.c
index f9b6023..6d7412a 100644
--- a/drivers/net/ethernet/chelsio/cxgb/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb/sge.c
@@ -1678,8 +1678,7 @@ static int t1_sge_tx(struct sk_buff *skb, struct adapter *adapter,
Index: linux-3.4/drivers/net/ethernet/chelsio/cxgb/sge.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/chelsio/cxgb/sge.c
+++ linux-3.4/drivers/net/ethernet/chelsio/cxgb/sge.c
@@ -1678,8 +1678,7 @@ static int t1_sge_tx(struct sk_buff *skb
struct cmdQ *q = &sge->cmdQ[qid];
unsigned int credits, pidx, genbit, count, use_sched_skb = 0;
@ -66,11 +66,11 @@ index f9b6023..6d7412a 100644
reclaim_completed_tx(sge, q);
diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c
index c27fb3d..4624278 100644
--- a/drivers/net/ethernet/neterion/s2io.c
+++ b/drivers/net/ethernet/neterion/s2io.c
@@ -4090,12 +4090,7 @@ static netdev_tx_t s2io_xmit(struct sk_buff *skb, struct net_device *dev)
Index: linux-3.4/drivers/net/ethernet/neterion/s2io.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/neterion/s2io.c
+++ linux-3.4/drivers/net/ethernet/neterion/s2io.c
@@ -4089,12 +4089,7 @@ static netdev_tx_t s2io_xmit(struct sk_b
[skb->priority & (MAX_TX_FIFOS - 1)];
fifo = &mac_control->fifos[queue];
@ -84,11 +84,11 @@ index c27fb3d..4624278 100644
if (sp->config.multiq) {
if (__netif_subqueue_stopped(dev, fifo->fifo_no)) {
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
index 43c7b25..c084bea 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
@@ -1937,10 +1937,9 @@ static int pch_gbe_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
Index: linux-3.4/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
+++ linux-3.4/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
@@ -2131,10 +2131,9 @@ static int pch_gbe_xmit_frame(struct sk_
adapter->stats.tx_length_errors++;
return NETDEV_TX_OK;
}
@ -102,11 +102,11 @@ index 43c7b25..c084bea 100644
if (unlikely(!PCH_GBE_DESC_UNUSED(tx_ring))) {
netif_stop_queue(netdev);
spin_unlock_irqrestore(&tx_ring->tx_lock, flags);
diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c
index 3a90af6..e2e930e 100644
--- a/drivers/net/ethernet/tehuti/tehuti.c
+++ b/drivers/net/ethernet/tehuti/tehuti.c
@@ -1605,13 +1605,8 @@ static netdev_tx_t bdx_tx_transmit(struct sk_buff *skb,
Index: linux-3.4/drivers/net/ethernet/tehuti/tehuti.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/tehuti/tehuti.c
+++ linux-3.4/drivers/net/ethernet/tehuti/tehuti.c
@@ -1606,13 +1606,8 @@ static netdev_tx_t bdx_tx_transmit(struc
unsigned long flags;
ENTER;
@ -122,11 +122,11 @@ index 3a90af6..e2e930e 100644
/* build tx descriptor */
BDX_ASSERT(f->m.wptr >= f->m.memsz); /* started with valid wptr */
diff --git a/drivers/net/rionet.c b/drivers/net/rionet.c
index 7145714..2a1ed18 100644
--- a/drivers/net/rionet.c
+++ b/drivers/net/rionet.c
@@ -176,11 +176,7 @@ static int rionet_start_xmit(struct sk_buff *skb, struct net_device *ndev)
Index: linux-3.4/drivers/net/rionet.c
===================================================================
--- linux-3.4.orig/drivers/net/rionet.c
+++ linux-3.4/drivers/net/rionet.c
@@ -176,11 +176,7 @@ static int rionet_start_xmit(struct sk_b
u16 destid;
unsigned long flags;
@ -139,6 +139,3 @@ index 7145714..2a1ed18 100644
if ((rnet->tx_cnt + 1) > RIONET_TX_RING_SIZE) {
netif_stop_queue(ndev);
--
1.7.10

23
debian/patches/features/all/rt/0101-drivers-net-gianfar-Make-RT-aware.patch → debian/patches/features/all/rt/drivers-net-gianfar-make-rt-aware.patch vendored
View File

@ -1,7 +1,6 @@
From f6a04ba9615eedce21fa38c8a9276c1c041f4535 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 1 Apr 2010 20:20:57 +0200
Subject: [PATCH 101/267] drivers: net: gianfar: Make RT aware
Subject: drivers: net: gianfar: Make RT aware
The adjust_link() disables interrupts before taking the queue
locks. On RT those locks are converted to "sleeping" locks and
@ -11,15 +10,16 @@ local_irq_save/restore_nort.
Reported-by: Xianghua Xiao <xiaoxianghua@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Xianghua Xiao <xiaoxianghua@gmail.com>
---
drivers/net/ethernet/freescale/gianfar.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index 83199fd..97d238c 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -1671,7 +1671,7 @@ void stop_gfar(struct net_device *dev)
Index: linux-3.4/drivers/net/ethernet/freescale/gianfar.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/freescale/gianfar.c
+++ linux-3.4/drivers/net/ethernet/freescale/gianfar.c
@@ -1643,7 +1643,7 @@ void stop_gfar(struct net_device *dev)
/* Lock it down */
@ -28,7 +28,7 @@ index 83199fd..97d238c 100644
lock_tx_qs(priv);
lock_rx_qs(priv);
@@ -1679,7 +1679,7 @@ void stop_gfar(struct net_device *dev)
@@ -1651,7 +1651,7 @@ void stop_gfar(struct net_device *dev)
unlock_rx_qs(priv);
unlock_tx_qs(priv);
@ -37,7 +37,7 @@ index 83199fd..97d238c 100644
/* Free the IRQs */
if (priv->device_flags & FSL_GIANFAR_DEV_HAS_MULTI_INTR) {
@@ -2949,7 +2949,7 @@ static void adjust_link(struct net_device *dev)
@@ -2948,7 +2948,7 @@ static void adjust_link(struct net_devic
struct phy_device *phydev = priv->phydev;
int new_state = 0;
@ -46,7 +46,7 @@ index 83199fd..97d238c 100644
lock_tx_qs(priv);
if (phydev->link) {
@@ -3016,7 +3016,7 @@ static void adjust_link(struct net_device *dev)
@@ -3015,7 +3015,7 @@ static void adjust_link(struct net_devic
if (new_state && netif_msg_link(priv))
phy_print_status(phydev);
unlock_tx_qs(priv);
@ -55,6 +55,3 @@ index 83199fd..97d238c 100644
}
/* Update the hash table based on the current list of multicast
--
1.7.10

18
debian/patches/features/all/rt/0040-drivers-net-tulip_remove_one-needs-to-call-pci_disab.patch → debian/patches/features/all/rt/drivers-net-tulip-add-missing-pci-disable.patch vendored
View File

@ -1,22 +1,21 @@
From 32a59ec7007ad587e8523ea3628b364c434d4c54 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:30:18 -0500
Subject: [PATCH 040/267] drivers/net: tulip_remove_one needs to call
pci_disable_device()
Subject: drivers/net: tulip_remove_one needs to call pci_disable_device()
Otherwise the device is not completely shut down.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/net/ethernet/dec/tulip/tulip_core.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c
index 9656dd0..ef7df09 100644
--- a/drivers/net/ethernet/dec/tulip/tulip_core.c
+++ b/drivers/net/ethernet/dec/tulip/tulip_core.c
@@ -1949,6 +1949,7 @@ static void __devexit tulip_remove_one (struct pci_dev *pdev)
Index: linux-3.4/drivers/net/ethernet/dec/tulip/tulip_core.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/dec/tulip/tulip_core.c
+++ linux-3.4/drivers/net/ethernet/dec/tulip/tulip_core.c
@@ -1946,6 +1946,7 @@ static void __devexit tulip_remove_one (
pci_iounmap(pdev, tp->base_addr);
free_netdev (dev);
pci_release_regions (pdev);
@ -24,6 +23,3 @@ index 9656dd0..ef7df09 100644
pci_set_drvdata (pdev, NULL);
/* pci_power_off (pdev, -1); */
--
1.7.10

21
debian/patches/features/all/rt/0100-drivers-net-vortex-fix-locking-issues.patch → debian/patches/features/all/rt/drivers-net-vortex-fix-locking-issues.patch vendored
View File

@ -1,7 +1,6 @@
From c9c1db28cb2aece6eeda56b86a309b2d564751b8 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 3 Jul 2009 08:30:00 -0500
Subject: [PATCH 100/267] drivers/net: vortex fix locking issues
Subject: drivers/net: vortex fix locking issues
Argh, cut and paste wasn't enough...
@ -17,15 +16,12 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
1 file changed, 4 insertions(+), 4 deletions(-)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
drivers/net/ethernet/3com/3c59x.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
index e0c5529..af31580 100644
--- a/drivers/net/ethernet/3com/3c59x.c
+++ b/drivers/net/ethernet/3com/3c59x.c
@@ -843,9 +843,9 @@ static void poll_vortex(struct net_device *dev)
Index: linux-3.4/drivers/net/ethernet/3com/3c59x.c
===================================================================
--- linux-3.4.orig/drivers/net/ethernet/3com/3c59x.c
+++ linux-3.4/drivers/net/ethernet/3com/3c59x.c
@@ -843,9 +843,9 @@ static void poll_vortex(struct net_devic
{
struct vortex_private *vp = netdev_priv(dev);
unsigned long flags;
@ -37,7 +33,7 @@ index e0c5529..af31580 100644
}
#endif
@@ -1921,12 +1921,12 @@ static void vortex_tx_timeout(struct net_device *dev)
@@ -1920,12 +1920,12 @@ static void vortex_tx_timeout(struct net
* Block interrupts because vortex_interrupt does a bare spin_lock()
*/
unsigned long flags;
@ -52,6 +48,3 @@ index e0c5529..af31580 100644
}
}
--
1.7.10

21
debian/patches/features/all/rt/0037-drivers-random-Reduce-preempt-disabled-region.patch → debian/patches/features/all/rt/drivers-random-reduce-preempt-disabled-region.patch vendored
View File

@ -1,21 +1,21 @@
From a908abba18460e32b36e6b183e77b6c7c1cceb94 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:30 -0500
Subject: [PATCH 037/267] drivers: random: Reduce preempt disabled region
Subject: drivers: random: Reduce preempt disabled region
No need to keep preemption disabled across the whole function.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/char/random.c | 9 +++++----
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/char/random.c b/drivers/char/random.c
index 6035ab8..786a856 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -633,8 +633,11 @@ static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
Index: linux-3.4/drivers/char/random.c
===================================================================
--- linux-3.4.orig/drivers/char/random.c
+++ linux-3.4/drivers/char/random.c
@@ -633,8 +633,11 @@ static void add_timer_randomness(struct
preempt_disable();
/* if over the trickle threshold, use only 1 in 4096 samples */
if (input_pool.entropy_count > trickle_thresh &&
@ -28,8 +28,8 @@ index 6035ab8..786a856 100644
+ preempt_enable();
sample.jiffies = jiffies;
sample.cycles = get_cycles();
@@ -676,8 +679,6 @@ static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
@@ -680,8 +683,6 @@ static void add_timer_randomness(struct
credit_entropy_bits(&input_pool,
min_t(int, fls(delta>>1), 11));
}
@ -38,6 +38,3 @@ index 6035ab8..786a856 100644
}
void add_input_randomness(unsigned int type, unsigned int code,
--
1.7.10

32
debian/patches/features/all/rt/0212-serial-8250-Call-flush_to_ldisc-when-the-irq-is-thre.patch → debian/patches/features/all/rt/drivers-serial-call-flush_to_ldisc-when-the-irq-is-t.patch vendored
View File

@ -1,20 +1,19 @@
From 4e5bbac0b9e6295553e4e4ddc234f1ccdb54f20d Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:30:01 -0500
Subject: [PATCH 212/267] serial: 8250: Call flush_to_ldisc when the irq is
threaded
Subject: serial: 8250: Call flush_to_ldisc when the irq is threaded
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
drivers/tty/serial/8250.c | 2 ++
drivers/tty/tty_buffer.c | 4 ++++
drivers/tty/serial/8250/8250.c | 2 ++
drivers/tty/tty_buffer.c | 4 ++++
2 files changed, 6 insertions(+)
diff --git a/drivers/tty/serial/8250.c b/drivers/tty/serial/8250.c
index 43f7f46..687b1ad 100644
--- a/drivers/tty/serial/8250.c
+++ b/drivers/tty/serial/8250.c
@@ -1631,12 +1631,14 @@ static irqreturn_t serial8250_interrupt(int irq, void *dev_id)
Index: linux-3.4/drivers/tty/serial/8250/8250.c
===================================================================
--- linux-3.4.orig/drivers/tty/serial/8250/8250.c
+++ linux-3.4/drivers/tty/serial/8250/8250.c
@@ -1584,12 +1584,14 @@ static irqreturn_t serial8250_interrupt(
l = l->next;
@ -29,11 +28,11 @@ index 43f7f46..687b1ad 100644
} while (l != end);
spin_unlock(&i->lock);
diff --git a/drivers/tty/tty_buffer.c b/drivers/tty/tty_buffer.c
index 6c9b7cd..a56c223 100644
--- a/drivers/tty/tty_buffer.c
+++ b/drivers/tty/tty_buffer.c
@@ -493,10 +493,14 @@ void tty_flip_buffer_push(struct tty_struct *tty)
Index: linux-3.4/drivers/tty/tty_buffer.c
===================================================================
--- linux-3.4.orig/drivers/tty/tty_buffer.c
+++ linux-3.4/drivers/tty/tty_buffer.c
@@ -493,10 +493,14 @@ void tty_flip_buffer_push(struct tty_str
tty->buf.tail->commit = tty->buf.tail->used;
spin_unlock_irqrestore(&tty->buf.lock, flags);
@ -48,6 +47,3 @@ index 6c9b7cd..a56c223 100644
}
EXPORT_SYMBOL(tty_flip_buffer_push);
--
1.7.10

44
debian/patches/features/all/rt/drivers-serial-cleanup-locking-for-rt.patch vendored Normal file
View File

@ -0,0 +1,44 @@
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:30:01 -0500
Subject: serial: 8250: Clean up the locking for -rt
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/tty/serial/8250/8250.c | 15 +++++----------
1 file changed, 5 insertions(+), 10 deletions(-)
Index: linux-3.4/drivers/tty/serial/8250/8250.c
===================================================================
--- linux-3.4.orig/drivers/tty/serial/8250/8250.c
+++ linux-3.4/drivers/tty/serial/8250/8250.c
@@ -2807,14 +2807,10 @@ serial8250_console_write(struct console
touch_nmi_watchdog();
- local_irq_save(flags);
- if (port->sysrq) {
- /* serial8250_handle_irq() already took the lock */
- locked = 0;
- } else if (oops_in_progress) {
- locked = spin_trylock(&port->lock);
- } else
- spin_lock(&port->lock);
+ if (port->sysrq || oops_in_progress)
+ locked = spin_trylock_irqsave(&port->lock, flags);
+ else
+ spin_lock_irqsave(&port->lock, flags);
/*
* First save the IER then disable the interrupts
@@ -2846,8 +2842,7 @@ serial8250_console_write(struct console
serial8250_modem_status(up);
if (locked)
- spin_unlock(&port->lock);
- local_irq_restore(flags);
+ spin_unlock_irqrestore(&port->lock, flags);
}
static int __init serial8250_console_setup(struct console *co, char *options)

40
debian/patches/features/all/rt/drivers-tty-fix-omap-lock-crap.patch vendored Normal file
View File

@ -0,0 +1,40 @@
Subject: drivers-tty-fix-omap-lock-crap.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Thu, 28 Jul 2011 13:32:57 +0200
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
drivers/tty/serial/omap-serial.c | 12 ++++--------
1 file changed, 4 insertions(+), 8 deletions(-)
Index: linux-3.4/drivers/tty/serial/omap-serial.c
===================================================================
--- linux-3.4.orig/drivers/tty/serial/omap-serial.c
+++ linux-3.4/drivers/tty/serial/omap-serial.c
@@ -1064,13 +1064,10 @@ serial_omap_console_write(struct console
pm_runtime_get_sync(&up->pdev->dev);
- local_irq_save(flags);
- if (up->port.sysrq)
- locked = 0;
- else if (oops_in_progress)
- locked = spin_trylock(&up->port.lock);
+ if (up->port.sysrq || oops_in_progress)
+ locked = spin_trylock_irqsave(&up->port.lock, flags);
else
- spin_lock(&up->port.lock);
+ spin_lock_irqsave(&up->port.lock, flags);
/*
* First save the IER then disable the interrupts
@@ -1099,8 +1096,7 @@ serial_omap_console_write(struct console
pm_runtime_mark_last_busy(&up->pdev->dev);
pm_runtime_put_autosuspend(&up->pdev->dev);
if (locked)
- spin_unlock(&up->port.lock);
- local_irq_restore(flags);
+ spin_unlock_irqrestore(&up->port.lock, flags);
}
static int __init

183
debian/patches/features/all/rt/0073-early-printk-consolidate.patch.patch → debian/patches/features/all/rt/early-printk-consolidate.patch vendored
View File

@ -1,7 +1,6 @@
From d6a466f7aef3ff53e864d1d0a8f5e93d61768714 Mon Sep 17 00:00:00 2001
Subject: early-printk-consolidate.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 23 Jul 2011 11:04:08 +0200
Subject: [PATCH 073/267] early-printk-consolidate.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
@ -19,14 +18,14 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arch/x86/kernel/early_printk.c | 21 ++-------------------
include/linux/console.h | 1 +
include/linux/printk.h | 5 +++++
kernel/printk.c | 27 ++++++++++++++++++++-------
15 files changed, 62 insertions(+), 110 deletions(-)
kernel/printk.c | 30 +++++++++++++++++++++++-------
15 files changed, 65 insertions(+), 110 deletions(-)
diff --git a/arch/arm/kernel/early_printk.c b/arch/arm/kernel/early_printk.c
index 85aa2b2..4307653 100644
--- a/arch/arm/kernel/early_printk.c
+++ b/arch/arm/kernel/early_printk.c
@@ -29,28 +29,17 @@ static void early_console_write(struct console *con, const char *s, unsigned n)
Index: linux-3.4/arch/arm/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/arm/kernel/early_printk.c
+++ linux-3.4/arch/arm/kernel/early_printk.c
@@ -29,28 +29,17 @@ static void early_console_write(struct c
early_write(s, n);
}
@ -58,11 +57,11 @@ index 85aa2b2..4307653 100644
return 0;
}
diff --git a/arch/blackfin/kernel/early_printk.c b/arch/blackfin/kernel/early_printk.c
index 84ed837..61fbd2d 100644
--- a/arch/blackfin/kernel/early_printk.c
+++ b/arch/blackfin/kernel/early_printk.c
@@ -25,8 +25,6 @@ extern struct console *bfin_earlyserial_init(unsigned int port,
Index: linux-3.4/arch/blackfin/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/blackfin/kernel/early_printk.c
+++ linux-3.4/arch/blackfin/kernel/early_printk.c
@@ -25,8 +25,6 @@ extern struct console *bfin_earlyserial_
extern struct console *bfin_jc_early_init(void);
#endif
@ -71,10 +70,10 @@ index 84ed837..61fbd2d 100644
/* Default console */
#define DEFAULT_PORT 0
#define DEFAULT_CFLAG CS8|B57600
diff --git a/arch/microblaze/kernel/early_printk.c b/arch/microblaze/kernel/early_printk.c
index d26d92d..0420624 100644
--- a/arch/microblaze/kernel/early_printk.c
+++ b/arch/microblaze/kernel/early_printk.c
Index: linux-3.4/arch/microblaze/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/microblaze/kernel/early_printk.c
+++ linux-3.4/arch/microblaze/kernel/early_printk.c
@@ -21,7 +21,6 @@
#include <asm/setup.h>
#include <asm/prom.h>
@ -83,7 +82,7 @@ index d26d92d..0420624 100644
static u32 base_addr;
#ifdef CONFIG_SERIAL_UARTLITE_CONSOLE
@@ -109,27 +108,11 @@ static struct console early_serial_uart16550_console = {
@@ -109,27 +108,11 @@ static struct console early_serial_uart1
};
#endif /* CONFIG_SERIAL_8250_CONSOLE */
@ -127,9 +126,9 @@ index d26d92d..0420624 100644
- if (!early_console_initialized || !early_console)
+ if (!early_console)
return;
printk(KERN_INFO "early_printk_console remaping from 0x%x to ",
printk(KERN_INFO "early_printk_console remapping from 0x%x to ",
base_addr);
@@ -179,9 +161,9 @@ void __init remap_early_printk(void)
@@ -195,9 +177,9 @@ void __init remap_early_printk(void)
void __init disable_early_printk(void)
{
@ -141,11 +140,11 @@ index d26d92d..0420624 100644
- early_console_initialized = 0;
+ early_console = NULL;
}
diff --git a/arch/mips/kernel/early_printk.c b/arch/mips/kernel/early_printk.c
index 9ae813e..973c995 100644
--- a/arch/mips/kernel/early_printk.c
+++ b/arch/mips/kernel/early_printk.c
@@ -25,20 +25,18 @@ early_console_write(struct console *con, const char *s, unsigned n)
Index: linux-3.4/arch/mips/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/mips/kernel/early_printk.c
+++ linux-3.4/arch/mips/kernel/early_printk.c
@@ -25,20 +25,18 @@ early_console_write(struct console *con,
}
}
@ -170,11 +169,11 @@ index 9ae813e..973c995 100644
- register_console(&early_console);
+ register_console(&early_console_prom);
}
diff --git a/arch/powerpc/kernel/udbg.c b/arch/powerpc/kernel/udbg.c
index 57fa2c0..1b9174d 100644
--- a/arch/powerpc/kernel/udbg.c
+++ b/arch/powerpc/kernel/udbg.c
@@ -182,15 +182,13 @@ static struct console udbg_console = {
Index: linux-3.4/arch/powerpc/kernel/udbg.c
===================================================================
--- linux-3.4.orig/arch/powerpc/kernel/udbg.c
+++ linux-3.4/arch/powerpc/kernel/udbg.c
@@ -179,15 +179,13 @@ static struct console udbg_console = {
.index = 0,
};
@ -191,7 +190,7 @@ index 57fa2c0..1b9174d 100644
return;
if (!udbg_putc)
@@ -200,7 +198,7 @@ void __init register_early_udbg_console(void)
@@ -197,7 +195,7 @@ void __init register_early_udbg_console(
printk(KERN_INFO "early console immortal !\n");
udbg_console.flags &= ~CON_BOOT;
}
@ -200,10 +199,10 @@ index 57fa2c0..1b9174d 100644
register_console(&udbg_console);
}
diff --git a/arch/sh/kernel/sh_bios.c b/arch/sh/kernel/sh_bios.c
index 47475cc..a5b51b9 100644
--- a/arch/sh/kernel/sh_bios.c
+++ b/arch/sh/kernel/sh_bios.c
Index: linux-3.4/arch/sh/kernel/sh_bios.c
===================================================================
--- linux-3.4.orig/arch/sh/kernel/sh_bios.c
+++ linux-3.4/arch/sh/kernel/sh_bios.c
@@ -144,8 +144,6 @@ static struct console bios_console = {
.index = -1,
};
@ -213,10 +212,10 @@ index 47475cc..a5b51b9 100644
static int __init setup_early_printk(char *buf)
{
int keep_early = 0;
diff --git a/arch/sparc/kernel/setup_32.c b/arch/sparc/kernel/setup_32.c
index fe1e3fc..e6475f0 100644
--- a/arch/sparc/kernel/setup_32.c
+++ b/arch/sparc/kernel/setup_32.c
Index: linux-3.4/arch/sparc/kernel/setup_32.c
===================================================================
--- linux-3.4.orig/arch/sparc/kernel/setup_32.c
+++ linux-3.4/arch/sparc/kernel/setup_32.c
@@ -221,6 +221,7 @@ void __init setup_arch(char **cmdline_p)
boot_flags_init(*cmdline_p);
@ -225,11 +224,11 @@ index fe1e3fc..e6475f0 100644
register_console(&prom_early_console);
/* Set sparc_cpu_model */
diff --git a/arch/sparc/kernel/setup_64.c b/arch/sparc/kernel/setup_64.c
index a854a1c..b85d039 100644
--- a/arch/sparc/kernel/setup_64.c
+++ b/arch/sparc/kernel/setup_64.c
@@ -487,6 +487,12 @@ static void __init init_sparc64_elf_hwcap(void)
Index: linux-3.4/arch/sparc/kernel/setup_64.c
===================================================================
--- linux-3.4.orig/arch/sparc/kernel/setup_64.c
+++ linux-3.4/arch/sparc/kernel/setup_64.c
@@ -487,6 +487,12 @@ static void __init init_sparc64_elf_hwca
popc_patch();
}
@ -251,11 +250,11 @@ index a854a1c..b85d039 100644
if (tlb_type == hypervisor)
printk("ARCH: SUN4V\n");
diff --git a/arch/tile/kernel/early_printk.c b/arch/tile/kernel/early_printk.c
index 493a0e6..ba2ac00 100644
--- a/arch/tile/kernel/early_printk.c
+++ b/arch/tile/kernel/early_printk.c
@@ -32,25 +32,8 @@ static struct console early_hv_console = {
Index: linux-3.4/arch/tile/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/tile/kernel/early_printk.c
+++ linux-3.4/arch/tile/kernel/early_printk.c
@@ -33,25 +33,8 @@ static struct console early_hv_console =
};
/* Direct interface for emergencies */
@ -281,7 +280,7 @@ index 493a0e6..ba2ac00 100644
void early_panic(const char *fmt, ...)
{
va_list ap;
@@ -68,14 +51,13 @@ static int __initdata keep_early;
@@ -69,14 +52,13 @@ static int __initdata keep_early;
static int __init setup_early_printk(char *str)
{
@ -297,7 +296,7 @@ index 493a0e6..ba2ac00 100644
register_console(early_console);
return 0;
@@ -84,12 +66,12 @@ static int __init setup_early_printk(char *str)
@@ -85,12 +67,12 @@ static int __init setup_early_printk(cha
void __init disable_early_printk(void)
{
early_console_complete = 1;
@ -312,7 +311,7 @@ index 493a0e6..ba2ac00 100644
} else {
early_printk("keeping early console\n");
}
@@ -97,7 +79,7 @@ void __init disable_early_printk(void)
@@ -98,7 +80,7 @@ void __init disable_early_printk(void)
void warn_early_printk(void)
{
@ -321,11 +320,11 @@ index 493a0e6..ba2ac00 100644
return;
early_printk("\
Machine shutting down before console output is fully initialized.\n\
diff --git a/arch/um/kernel/early_printk.c b/arch/um/kernel/early_printk.c
index ec649bf..183060f 100644
--- a/arch/um/kernel/early_printk.c
+++ b/arch/um/kernel/early_printk.c
@@ -16,7 +16,7 @@ static void early_console_write(struct console *con, const char *s, unsigned int
Index: linux-3.4/arch/um/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/um/kernel/early_printk.c
+++ linux-3.4/arch/um/kernel/early_printk.c
@@ -16,7 +16,7 @@ static void early_console_write(struct c
um_early_printk(s, n);
}
@ -347,11 +346,11 @@ index ec649bf..183060f 100644
return 0;
}
diff --git a/arch/unicore32/kernel/early_printk.c b/arch/unicore32/kernel/early_printk.c
index 3922255..9be0d5d 100644
--- a/arch/unicore32/kernel/early_printk.c
+++ b/arch/unicore32/kernel/early_printk.c
@@ -33,21 +33,17 @@ static struct console early_ocd_console = {
Index: linux-3.4/arch/unicore32/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/unicore32/kernel/early_printk.c
+++ linux-3.4/arch/unicore32/kernel/early_printk.c
@@ -33,21 +33,17 @@ static struct console early_ocd_console
.index = -1,
};
@ -377,11 +376,11 @@ index 3922255..9be0d5d 100644
if (keep_early)
early_console->flags &= ~CON_BOOT;
diff --git a/arch/x86/kernel/early_printk.c b/arch/x86/kernel/early_printk.c
index cd28a35..5f3d9c5 100644
--- a/arch/x86/kernel/early_printk.c
+++ b/arch/x86/kernel/early_printk.c
@@ -169,25 +169,9 @@ static struct console early_serial_console = {
Index: linux-3.4/arch/x86/kernel/early_printk.c
===================================================================
--- linux-3.4.orig/arch/x86/kernel/early_printk.c
+++ linux-3.4/arch/x86/kernel/early_printk.c
@@ -169,25 +169,9 @@ static struct console early_serial_conso
.index = -1,
};
@ -408,7 +407,7 @@ index cd28a35..5f3d9c5 100644
printk(KERN_CRIT "ERROR: earlyprintk= %s already used\n",
con->name);
return;
@@ -207,9 +191,8 @@ static int __init setup_early_printk(char *buf)
@@ -207,9 +191,8 @@ static int __init setup_early_printk(cha
if (!buf)
return 0;
@ -419,10 +418,10 @@ index cd28a35..5f3d9c5 100644
keep = (strstr(buf, "keep") != NULL);
diff --git a/include/linux/console.h b/include/linux/console.h
index 7453cfd..e5b5dc0 100644
--- a/include/linux/console.h
+++ b/include/linux/console.h
Index: linux-3.4/include/linux/console.h
===================================================================
--- linux-3.4.orig/include/linux/console.h
+++ linux-3.4/include/linux/console.h
@@ -133,6 +133,7 @@ struct console {
for (con = console_drivers; con != NULL; con = con->next)
@ -431,10 +430,10 @@ index 7453cfd..e5b5dc0 100644
extern int add_preferred_console(char *name, int idx, char *options);
extern int update_console_cmdline(char *name, int idx, char *name_new, int idx_new, char *options);
diff --git a/include/linux/printk.h b/include/linux/printk.h
index f0e22f7..be10455 100644
--- a/include/linux/printk.h
+++ b/include/linux/printk.h
Index: linux-3.4/include/linux/printk.h
===================================================================
--- linux-3.4.orig/include/linux/printk.h
+++ linux-3.4/include/linux/printk.h
@@ -88,8 +88,13 @@ int no_printk(const char *fmt, ...)
return 0;
}
@ -449,13 +448,13 @@ index f0e22f7..be10455 100644
extern int printk_needs_cpu(int cpu);
extern void printk_tick(void);
diff --git a/kernel/printk.c b/kernel/printk.c
index 7982a0a..b7aa50e 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -44,13 +44,6 @@
#include <asm/uaccess.h>
Index: linux-3.4/kernel/printk.c
===================================================================
--- linux-3.4.orig/kernel/printk.c
+++ linux-3.4/kernel/printk.c
@@ -47,13 +47,6 @@
#define CREATE_TRACE_POINTS
#include <trace/events/printk.h>
-/*
- * Architectures can override it:
@ -467,7 +466,7 @@ index 7982a0a..b7aa50e 100644
#define __LOG_BUF_LEN (1 << CONFIG_LOG_BUF_SHIFT)
/* printk's without a loglevel use this.. */
@@ -521,6 +514,26 @@ static void __call_console_drivers(unsigned start, unsigned end)
@@ -524,6 +517,29 @@ static void __call_console_drivers(unsig
}
}
@ -476,24 +475,24 @@ index 7982a0a..b7aa50e 100644
+
+static void early_vprintk(const char *fmt, va_list ap)
+{
+ char buf[512];
+ int n = vscnprintf(buf, sizeof(buf), fmt, ap);
+ if (early_console)
+ if (early_console) {
+ char buf[512];
+ int n = vscnprintf(buf, sizeof(buf), fmt, ap);
+
+ early_console->write(early_console, buf, n);
+ }
+}
+
+asmlinkage void early_printk(const char *fmt, ...)
+{
+ va_list ap;
+
+ va_start(ap, fmt);
+ early_vprintk(fmt, ap);
+ va_end(ap);
+}
+#endif
+
static int __read_mostly ignore_loglevel;
static bool __read_mostly ignore_loglevel;
static int __init ignore_loglevel_setup(char *str)
--
1.7.10

18
debian/patches/features/all/rt/0224-epoll.patch.patch → debian/patches/features/all/rt/epoll-use-get-cpu-light.patch vendored
View File

@ -1,18 +1,17 @@
From b567a9a48741b5c6a271e478cc11149308ecd9b2 Mon Sep 17 00:00:00 2001
Subject: epoll.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 8 Jul 2011 16:35:35 +0200
Subject: [PATCH 224/267] epoll.patch
Date: Fri, 08 Jul 2011 16:35:35 +0200
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
fs/eventpoll.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index ea54cde..34294b9 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -464,12 +464,12 @@ static int ep_poll_wakeup_proc(void *priv, void *cookie, int call_nests)
Index: linux-3.4/fs/eventpoll.c
===================================================================
--- linux-3.4.orig/fs/eventpoll.c
+++ linux-3.4/fs/eventpoll.c
@@ -488,12 +488,12 @@ static int ep_poll_wakeup_proc(void *pri
*/
static void ep_poll_safewake(wait_queue_head_t *wq)
{
@ -27,6 +26,3 @@ index ea54cde..34294b9 100644
}
static void ep_remove_wait_queue(struct eppoll_entry *pwq)
--
1.7.10

18
debian/patches/features/all/rt/0049-filemap-fix-up.patch.patch → debian/patches/features/all/rt/filemap-fix-up.patch vendored
View File

@ -1,7 +1,6 @@
From 90e4703e2535c334bf845c719179075440e4fdca Mon Sep 17 00:00:00 2001
Subject: filemap-fix-up.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 17 Jun 2011 18:56:24 +0200
Subject: [PATCH 049/267] filemap-fix-up.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Wrecked-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
@ -10,19 +9,16 @@ Link: http://lkml.kernel.org/n/tip-m6yuzd6ul717hlnl2gj6p3ou@git.kernel.org
mm/filemap.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/filemap.c b/mm/filemap.c
index 03c5b0e..4a30d36 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2044,7 +2044,7 @@ size_t iov_iter_copy_from_user_atomic(struct page *page,
Index: linux-3.4/mm/filemap.c
===================================================================
--- linux-3.4.orig/mm/filemap.c
+++ linux-3.4/mm/filemap.c
@@ -2036,7 +2036,7 @@ size_t iov_iter_copy_from_user_atomic(st
char *kaddr;
size_t copied;
- BUG_ON(!in_atomic());
+ BUG_ON(!pagefault_disabled());
kaddr = kmap_atomic(page, KM_USER0);
kaddr = kmap_atomic(page);
if (likely(i->nr_segs == 1)) {
int left;
--
1.7.10

65
debian/patches/features/all/rt/fix-printk-flush-of-messages-1.patch vendored Normal file
View File

@ -0,0 +1,65 @@
Subject: fix printk flush of messages
From: Frank Rowand <frank.rowand@am.sony.com>
Date: Wed, 16 May 2012 18:09:36 -0700
Updates console-make-rt-friendly.patch
#ifdef CONFIG_PREEMPT_RT_FULL, printk() output is never flushed by
printk() because:
# some liberties taken in this pseudo-code to make it easier to follow
printk()
vprintk()
raw_spin_lock(&logbuf_lock)
# increment preempt_count():
preempt_disable()
result = console_trylock_for_printk()
retval = 0
# lock will always be false, because preempt_count() will be >= 1
lock = ... && !preempt_count()
if (lock)
retval = 1
return retval
# result will always be false since lock will always be false
if (result)
console_unlock()
# this is where the printk() output would be flushed
On system boot some printk() output is flushed because register_console()
and tty_open() call console_unlock().
This change also fixes the problem that was previously fixed by
preempt-rt-allow-immediate-magic-sysrq-output-for-preempt_rt_full.patch
Signed-off-by: Frank Rowand <frank.rowand@am.sony.com>
Cc: Frank <Frank_Rowand@sonyusa.com>
Link: http://lkml.kernel.org/r/4FB44FD0.4090800@am.sony.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/printk.c | 2 1 + 1 - 0 !
1 file changed, 1 insertion(+), 1 deletion(-)
Index: b/kernel/printk.c
===================================================================
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -847,7 +847,7 @@ static int console_trylock_for_printk(un
int retval = 0, wake = 0;
#ifdef CONFIG_PREEMPT_RT_FULL
int lock = !early_boot_irqs_disabled && !irqs_disabled_flags(flags) &&
- !preempt_count();
+ (preempt_count() <= 1);
#else
int lock = 1;
#endif
--
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

52
debian/patches/features/all/rt/0065-x86-Do-not-disable-preemption-in-int3-on-32bit.patch → debian/patches/features/all/rt/fix-rt-int3-x86_32-3.2-rt.patch vendored
View File

@ -1,7 +1,5 @@
From 31561541540fcec25c731c360a66fab17d50fa33 Mon Sep 17 00:00:00 2001
Subject: x86: Do not disable preemption in int3 on 32bit
From: Steven Rostedt <rostedt@goodmis.org>
Date: Tue, 10 Apr 2012 14:33:57 -0400
Subject: [PATCH 065/267] x86: Do not disable preemption in int3 on 32bit
Preemption must be disabled before enabling interrupts in do_trap
on x86_64 because the stack in use for int3 and debug is a per CPU
@ -19,15 +17,16 @@ when the stack is on the IST.
Cc: stable-rt@vger.kernel.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/kernel/traps.c | 32 +++++++++++++++++++++++---------
1 file changed, 23 insertions(+), 9 deletions(-)
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 31d9d0f..cc88aec 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -87,9 +87,21 @@ static inline void conditional_sti(struct pt_regs *regs)
Index: linux-3.4/arch/x86/kernel/traps.c
===================================================================
--- linux-3.4.orig/arch/x86/kernel/traps.c
+++ linux-3.4/arch/x86/kernel/traps.c
@@ -87,9 +87,21 @@ static inline void conditional_sti(struc
local_irq_enable();
}
@ -50,7 +49,7 @@ index 31d9d0f..cc88aec 100644
if (regs->flags & X86_EFLAGS_IF)
local_irq_enable();
}
@@ -100,11 +112,13 @@ static inline void conditional_cli(struct pt_regs *regs)
@@ -100,11 +112,13 @@ static inline void conditional_cli(struc
local_irq_disable();
}
@ -65,54 +64,51 @@ index 31d9d0f..cc88aec 100644
}
static void __kprobes
@@ -222,9 +236,9 @@ dotraplinkage void do_stack_segment(struct pt_regs *regs, long error_code)
@@ -226,9 +240,9 @@ dotraplinkage void do_stack_segment(stru
if (notify_die(DIE_TRAP, "stack segment", regs, error_code,
12, SIGBUS) == NOTIFY_STOP)
X86_TRAP_SS, SIGBUS) == NOTIFY_STOP)
return;
- preempt_conditional_sti(regs);
+ conditional_sti_ist(regs);
do_trap(12, SIGBUS, "stack segment", regs, error_code, NULL);
do_trap(X86_TRAP_SS, SIGBUS, "stack segment", regs, error_code, NULL);
- preempt_conditional_cli(regs);
+ conditional_cli_ist(regs);
}
dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
@@ -316,9 +330,9 @@ dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code)
return;
#endif
@@ -320,9 +334,9 @@ dotraplinkage void __kprobes do_int3(str
* as we may switch to the interrupt stack.
*/
debug_stack_usage_inc();
- preempt_conditional_sti(regs);
+ conditional_sti_ist(regs);
do_trap(3, SIGTRAP, "int3", regs, error_code, NULL);
do_trap(X86_TRAP_BP, SIGTRAP, "int3", regs, error_code, NULL);
- preempt_conditional_cli(regs);
+ conditional_cli_ist(regs);
debug_stack_usage_dec();
}
#ifdef CONFIG_X86_64
@@ -412,12 +426,12 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
return;
@@ -423,12 +437,12 @@ dotraplinkage void __kprobes do_debug(st
debug_stack_usage_inc();
/* It's safe to allow irq's after DR6 has been saved */
- preempt_conditional_sti(regs);
+ conditional_sti_ist(regs);
if (regs->flags & X86_VM_MASK) {
handle_vm86_trap((struct kernel_vm86_regs *) regs,
error_code, 1);
handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code,
X86_TRAP_DB);
- preempt_conditional_cli(regs);
+ conditional_cli_ist(regs);
debug_stack_usage_dec();
return;
}
@@ -436,7 +450,7 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
@@ -448,7 +462,7 @@ dotraplinkage void __kprobes do_debug(st
si_code = get_si_code(tsk->thread.debugreg6);
if (tsk->thread.debugreg6 & (DR_STEP | DR_TRAP_BITS) || user_icebp)
send_sigtrap(tsk, regs, error_code, si_code);
- preempt_conditional_cli(regs);
+ conditional_cli_ist(regs);
debug_stack_usage_dec();
return;
}
--
1.7.10

26
debian/patches/features/all/rt/0217-fs-block-rt-support.patch.patch → debian/patches/features/all/rt/fs-block-rt-support.patch vendored
View File

@ -1,7 +1,6 @@
From cfc9b78fd05098f317db725a3054e58d16e0da5e Mon Sep 17 00:00:00 2001
Subject: fs-block-rt-support.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 14 Jun 2011 17:05:09 +0200
Subject: [PATCH 217/267] fs-block-rt-support.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
@ -9,11 +8,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
fs/file.c | 4 ++--
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/block/blk-core.c b/block/blk-core.c
index 7366ad4..ca732c0 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -235,7 +235,7 @@ EXPORT_SYMBOL(blk_delay_queue);
Index: linux-3.4/block/blk-core.c
===================================================================
--- linux-3.4.orig/block/blk-core.c
+++ linux-3.4/block/blk-core.c
@@ -237,7 +237,7 @@ EXPORT_SYMBOL(blk_delay_queue);
**/
void blk_start_queue(struct request_queue *q)
{
@ -22,11 +21,11 @@ index 7366ad4..ca732c0 100644
queue_flag_clear(QUEUE_FLAG_STOPPED, q);
__blk_run_queue(q);
diff --git a/fs/file.c b/fs/file.c
index 375472d..fd03258 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -105,14 +105,14 @@ void free_fdtable_rcu(struct rcu_head *rcu)
Index: linux-3.4/fs/file.c
===================================================================
--- linux-3.4.orig/fs/file.c
+++ linux-3.4/fs/file.c
@@ -105,14 +105,14 @@ void free_fdtable_rcu(struct rcu_head *r
kfree(fdt->open_fds);
kfree(fdt);
} else {
@ -43,6 +42,3 @@ index 375472d..fd03258 100644
}
}
--
1.7.10

68
debian/patches/features/all/rt/0262-fs-dcache-Use-cpu_chill-in-trylock-loops.patch → debian/patches/features/all/rt/fs-dcache-use-cpu-chill-in-trylock-loops.patch vendored
View File

@ -1,7 +1,6 @@
From 1f3745ebd019755db68c544e5c9792988feebf52 Mon Sep 17 00:00:00 2001
Subject: fs: dcache: Use cpu_chill() in trylock loops
From: Thomas Gleixner <tglx@linutronix.de>
Date: Wed, 7 Mar 2012 21:00:34 +0100
Subject: [PATCH 262/267] fs: dcache: Use cpu_chill() in trylock loops
Date: Wed, 07 Mar 2012 21:00:34 +0100
Retry loops on RT might loop forever when the modifying side was
preempted. Use cpu_chill() instead of cpu_relax() to let the system
@ -16,10 +15,10 @@ Cc: stable-rt@vger.kernel.org
fs/namespace.c | 3 ++-
4 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 308a98b..67ed186 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
Index: linux-3.4/fs/autofs4/autofs_i.h
===================================================================
--- linux-3.4.orig/fs/autofs4/autofs_i.h
+++ linux-3.4/fs/autofs4/autofs_i.h
@@ -34,6 +34,7 @@
#include <linux/sched.h>
#include <linux/mount.h>
@ -28,11 +27,11 @@ index 308a98b..67ed186 100644
#include <asm/current.h>
#include <asm/uaccess.h>
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index 450f529..43afbb0 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -170,7 +170,7 @@ again:
Index: linux-3.4/fs/autofs4/expire.c
===================================================================
--- linux-3.4.orig/fs/autofs4/expire.c
+++ linux-3.4/fs/autofs4/expire.c
@@ -171,7 +171,7 @@ again:
parent = p->d_parent;
if (!spin_trylock(&parent->d_lock)) {
spin_unlock(&p->d_lock);
@ -41,19 +40,19 @@ index 450f529..43afbb0 100644
goto relock;
}
spin_unlock(&p->d_lock);
diff --git a/fs/dcache.c b/fs/dcache.c
index eb723d3..d47b266 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
Index: linux-3.4/fs/dcache.c
===================================================================
--- linux-3.4.orig/fs/dcache.c
+++ linux-3.4/fs/dcache.c
@@ -37,6 +37,7 @@
#include <linux/rculist_bl.h>
#include <linux/prefetch.h>
#include <linux/ratelimit.h>
+#include <linux/delay.h>
#include "internal.h"
#include "mount.h"
/*
@@ -410,7 +411,7 @@ static inline struct dentry *dentry_kill(struct dentry *dentry, int ref)
@@ -454,7 +455,7 @@ static inline struct dentry *dentry_kill
if (inode && !spin_trylock(&inode->i_lock)) {
relock:
spin_unlock(&dentry->d_lock);
@ -62,7 +61,7 @@ index eb723d3..d47b266 100644
return dentry; /* try again with same dentry */
}
if (IS_ROOT(dentry))
@@ -796,7 +797,7 @@ relock:
@@ -840,7 +841,7 @@ relock:
if (!spin_trylock(&dentry->d_lock)) {
spin_unlock(&dcache_lru_lock);
@ -71,7 +70,7 @@ index eb723d3..d47b266 100644
goto relock;
}
@@ -1975,7 +1976,7 @@ again:
@@ -2022,7 +2023,7 @@ again:
if (dentry->d_count == 1) {
if (inode && !spin_trylock(&inode->i_lock)) {
spin_unlock(&dentry->d_lock);
@ -80,27 +79,24 @@ index eb723d3..d47b266 100644
goto again;
}
dentry->d_flags &= ~DCACHE_CANT_MOUNT;
diff --git a/fs/namespace.c b/fs/namespace.c
index 63fd6d7..311264b 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -31,6 +31,7 @@
#include <linux/idr.h>
#include <linux/fs_struct.h>
#include <linux/fsnotify.h>
Index: linux-3.4/fs/namespace.c
===================================================================
--- linux-3.4.orig/fs/namespace.c
+++ linux-3.4/fs/namespace.c
@@ -20,6 +20,7 @@
#include <linux/fs_struct.h> /* get_fs_root et.al. */
#include <linux/fsnotify.h> /* fsnotify_vfsmount_delete */
#include <linux/uaccess.h>
+#include <linux/delay.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
#include "pnode.h"
@@ -346,7 +347,7 @@ int mnt_want_write(struct vfsmount *mnt)
*/
while (mnt->mnt_flags & MNT_WRITE_HOLD) {
#include "internal.h"
@@ -315,7 +316,7 @@ int mnt_want_write(struct vfsmount *m)
smp_mb();
while (mnt->mnt.mnt_flags & MNT_WRITE_HOLD) {
preempt_enable();
- cpu_relax();
+ cpu_chill();
preempt_disable();
}
/*
--
1.7.10

31
debian/patches/features/all/rt/0094-fs-jbd-jbd2-Make-state-lock-and-journal-head-lock-rt.patch → debian/patches/features/all/rt/fs-jbd-replace-bh_state-lock.patch vendored
View File

@ -1,8 +1,6 @@
From c9293d9a9b626e39108b276a74a6d4794374a973 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 18 Mar 2011 10:11:25 +0100
Subject: [PATCH 094/267] fs: jbd/jbd2: Make state lock and journal head lock
rt safe
Subject: fs: jbd/jbd2: Make state lock and journal head lock rt safe
bit_spin_locks break under RT.
@ -14,15 +12,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
include/linux/buffer_head.h | 10 ++++++++++
include/linux/jbd_common.h | 24 ++++++++++++++++++++++++
2 files changed, 34 insertions(+)
---
include/linux/buffer_head.h | 10 ++++++++++
include/linux/jbd_common.h | 24 ++++++++++++++++++++++++
2 files changed, 34 insertions(+)
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index 5c16cf1..3f8e27b 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
Index: linux-3.4/include/linux/buffer_head.h
===================================================================
--- linux-3.4.orig/include/linux/buffer_head.h
+++ linux-3.4/include/linux/buffer_head.h
@@ -74,6 +74,11 @@ struct buffer_head {
atomic_t b_count; /* users using this buffer_head */
#ifdef CONFIG_PREEMPT_RT_BASE
@ -35,7 +29,7 @@ index 5c16cf1..3f8e27b 100644
#endif
};
@@ -105,6 +110,11 @@ static inline void buffer_head_init_locks(struct buffer_head *bh)
@@ -105,6 +110,11 @@ static inline void buffer_head_init_lock
{
#ifdef CONFIG_PREEMPT_RT_BASE
spin_lock_init(&bh->b_uptodate_lock);
@ -47,11 +41,11 @@ index 5c16cf1..3f8e27b 100644
#endif
}
diff --git a/include/linux/jbd_common.h b/include/linux/jbd_common.h
index 6230f85..11c313e 100644
--- a/include/linux/jbd_common.h
+++ b/include/linux/jbd_common.h
@@ -37,32 +37,56 @@ static inline struct journal_head *bh2jh(struct buffer_head *bh)
Index: linux-3.4/include/linux/jbd_common.h
===================================================================
--- linux-3.4.orig/include/linux/jbd_common.h
+++ linux-3.4/include/linux/jbd_common.h
@@ -37,32 +37,56 @@ static inline struct journal_head *bh2jh
static inline void jbd_lock_bh_state(struct buffer_head *bh)
{
@ -108,6 +102,3 @@ index 6230f85..11c313e 100644
}
#endif
--
1.7.10

32
debian/patches/features/all/rt/fs-namespace-preemption-fix.patch vendored Normal file
View File

@ -0,0 +1,32 @@
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 19 Jul 2009 08:44:27 -0500
Subject: fs: namespace preemption fix
On RT we cannot loop with preemption disabled here as
mnt_make_readonly() might have been preempted. We can safely enable
preemption while waiting for MNT_WRITE_HOLD to be cleared. Safe on !RT
as well.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
fs/namespace.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
Index: linux-3.4/fs/namespace.c
===================================================================
--- linux-3.4.orig/fs/namespace.c
+++ linux-3.4/fs/namespace.c
@@ -313,8 +313,11 @@ int mnt_want_write(struct vfsmount *m)
* incremented count after it has set MNT_WRITE_HOLD.
*/
smp_mb();
- while (mnt->mnt.mnt_flags & MNT_WRITE_HOLD)
+ while (mnt->mnt.mnt_flags & MNT_WRITE_HOLD) {
+ preempt_enable();
cpu_relax();
+ preempt_disable();
+ }
/*
* After the slowpath clears MNT_WRITE_HOLD, mnt_is_readonly will
* be set to match its requirements. So we must not load that until

21
debian/patches/features/all/rt/0218-fs-ntfs-disable-interrupt-only-on-RT.patch → debian/patches/features/all/rt/fs-ntfs-disable-interrupt-non-rt.patch vendored
View File

@ -1,7 +1,6 @@
From 9091cea711af39936f8b1378fdb38f616e1dd15d Mon Sep 17 00:00:00 2001
From: Mike Galbraith <efault@gmx.de>
Date: Fri, 3 Jul 2009 08:44:12 -0500
Subject: [PATCH 218/267] fs: ntfs: disable interrupt only on !RT
Subject: fs: ntfs: disable interrupt only on !RT
On Sat, 2007-10-27 at 11:44 +0200, Ingo Molnar wrote:
> * Nick Piggin <nickpiggin@yahoo.com.au> wrote:
@ -35,30 +34,28 @@ instead?
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
fs/ntfs/aops.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
index 7fb7f1b..4c8095c 100644
--- a/fs/ntfs/aops.c
+++ b/fs/ntfs/aops.c
@@ -144,13 +144,13 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
Index: linux-3.4/fs/ntfs/aops.c
===================================================================
--- linux-3.4.orig/fs/ntfs/aops.c
+++ linux-3.4/fs/ntfs/aops.c
@@ -144,13 +144,13 @@ static void ntfs_end_buffer_async_read(s
recs = PAGE_CACHE_SIZE / rec_size;
/* Should have been verified before we got here... */
BUG_ON(!recs);
- local_irq_save(flags);
+ local_irq_save_nort(flags);
kaddr = kmap_atomic(page, KM_BIO_SRC_IRQ);
kaddr = kmap_atomic(page);
for (i = 0; i < recs; i++)
post_read_mst_fixup((NTFS_RECORD*)(kaddr +
i * rec_size), rec_size);
kunmap_atomic(kaddr, KM_BIO_SRC_IRQ);
kunmap_atomic(kaddr);
- local_irq_restore(flags);
+ local_irq_restore_nort(flags);
flush_dcache_page(page);
if (likely(page_uptodate && !PageError(page)))
SetPageUptodate(page);
--
1.7.10

48
debian/patches/features/all/rt/0093-buffer_head-Replace-bh_uptodate_lock-for-rt.patch → debian/patches/features/all/rt/fs-replace-bh_uptodate_lock-for-rt.patch vendored
View File

@ -1,7 +1,6 @@
From 0d6994d78eab6e6ebe324aa490ad10ddd2fc4e31 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 18 Mar 2011 09:18:52 +0100
Subject: [PATCH 093/267] buffer_head: Replace bh_uptodate_lock for -rt
Subject: buffer_head: Replace bh_uptodate_lock for -rt
Wrap the bit_spin_lock calls into a separate inline and add the RT
replacements with a real spinlock.
@ -13,11 +12,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
include/linux/buffer_head.h | 34 ++++++++++++++++++++++++++++++++++
3 files changed, 44 insertions(+), 21 deletions(-)
diff --git a/fs/buffer.c b/fs/buffer.c
index 19d8eb7..32c4680 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -331,8 +331,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
Index: linux-3.4/fs/buffer.c
===================================================================
--- linux-3.4.orig/fs/buffer.c
+++ linux-3.4/fs/buffer.c
@@ -281,8 +281,7 @@ static void end_buffer_async_read(struct
* decide that the page is now completely done.
*/
first = page_buffers(page);
@ -27,7 +26,7 @@ index 19d8eb7..32c4680 100644
clear_buffer_async_read(bh);
unlock_buffer(bh);
tmp = bh;
@@ -345,8 +344,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
@@ -295,8 +294,7 @@ static void end_buffer_async_read(struct
}
tmp = tmp->b_this_page;
} while (tmp != bh);
@ -37,7 +36,7 @@ index 19d8eb7..32c4680 100644
/*
* If none of the buffers had errors and they are all
@@ -358,9 +356,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
@@ -308,9 +306,7 @@ static void end_buffer_async_read(struct
return;
still_busy:
@ -48,7 +47,7 @@ index 19d8eb7..32c4680 100644
}
/*
@@ -394,8 +390,7 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
@@ -344,8 +340,7 @@ void end_buffer_async_write(struct buffe
}
first = page_buffers(page);
@ -58,7 +57,7 @@ index 19d8eb7..32c4680 100644
clear_buffer_async_write(bh);
unlock_buffer(bh);
@@ -407,15 +402,12 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
@@ -357,15 +352,12 @@ void end_buffer_async_write(struct buffe
}
tmp = tmp->b_this_page;
}
@ -76,7 +75,7 @@ index 19d8eb7..32c4680 100644
}
EXPORT_SYMBOL(end_buffer_async_write);
@@ -3223,6 +3215,7 @@ struct buffer_head *alloc_buffer_head(gfp_t gfp_flags)
@@ -3185,6 +3177,7 @@ struct buffer_head *alloc_buffer_head(gf
struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags);
if (ret) {
INIT_LIST_HEAD(&ret->b_assoc_buffers);
@ -84,11 +83,11 @@ index 19d8eb7..32c4680 100644
preempt_disable();
__this_cpu_inc(bh_accounting.nr);
recalc_bh_state();
diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
index 0b1e885b..7fb7f1b 100644
--- a/fs/ntfs/aops.c
+++ b/fs/ntfs/aops.c
@@ -108,8 +108,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
Index: linux-3.4/fs/ntfs/aops.c
===================================================================
--- linux-3.4.orig/fs/ntfs/aops.c
+++ linux-3.4/fs/ntfs/aops.c
@@ -108,8 +108,7 @@ static void ntfs_end_buffer_async_read(s
"0x%llx.", (unsigned long long)bh->b_blocknr);
}
first = page_buffers(page);
@ -98,7 +97,7 @@ index 0b1e885b..7fb7f1b 100644
clear_buffer_async_read(bh);
unlock_buffer(bh);
tmp = bh;
@@ -124,8 +123,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
@@ -124,8 +123,7 @@ static void ntfs_end_buffer_async_read(s
}
tmp = tmp->b_this_page;
} while (tmp != bh);
@ -108,7 +107,7 @@ index 0b1e885b..7fb7f1b 100644
/*
* If none of the buffers had errors then we can set the page uptodate,
* but we first have to perform the post read mst fixups, if the
@@ -160,9 +158,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
@@ -160,9 +158,7 @@ static void ntfs_end_buffer_async_read(s
unlock_page(page);
return;
still_busy:
@ -119,10 +118,10 @@ index 0b1e885b..7fb7f1b 100644
}
/**
diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
index 458f497..5c16cf1 100644
--- a/include/linux/buffer_head.h
+++ b/include/linux/buffer_head.h
Index: linux-3.4/include/linux/buffer_head.h
===================================================================
--- linux-3.4.orig/include/linux/buffer_head.h
+++ linux-3.4/include/linux/buffer_head.h
@@ -72,8 +72,42 @@ struct buffer_head {
struct address_space *b_assoc_map; /* mapping this buffer is
associated with */
@ -166,6 +165,3 @@ index 458f497..5c16cf1 100644
/*
* macro tricks to expand the set_buffer_foo(), clear_buffer_foo()
* and buffer_foo() functions.
--
1.7.10

36
debian/patches/features/all/rt/0175-ftrace-crap.patch.patch → debian/patches/features/all/rt/ftrace-crap.patch vendored
View File

@ -1,7 +1,6 @@
From 5af317e8a8abd8e488ee2a318229efc3177c81fc Mon Sep 17 00:00:00 2001
Subject: ftrace-crap.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 9 Sep 2011 16:55:53 +0200
Subject: [PATCH 175/267] ftrace-crap.patch
Date: Fri, 09 Sep 2011 16:55:53 +0200
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
@ -9,11 +8,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kernel/trace/trace.h | 1 -
2 files changed, 24 insertions(+), 3 deletions(-)
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index ff03353..f4de7ab 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -359,11 +359,13 @@ static DECLARE_DELAYED_WORK(wakeup_work, wakeup_work_handler);
Index: linux-3.4/kernel/trace/trace.c
===================================================================
--- linux-3.4.orig/kernel/trace/trace.c
+++ linux-3.4/kernel/trace/trace.c
@@ -414,11 +414,13 @@ EXPORT_SYMBOL_GPL(tracing_is_on);
*/
void trace_wake_up(void)
{
@ -27,7 +26,7 @@ index ff03353..f4de7ab 100644
}
static int __init set_buf_size(char *str)
@@ -719,6 +721,12 @@ update_max_tr_single(struct trace_array *tr, struct task_struct *tsk, int cpu)
@@ -775,6 +777,12 @@ update_max_tr_single(struct trace_array
}
#endif /* CONFIG_TRACER_MAX_TRACE */
@ -40,7 +39,7 @@ index ff03353..f4de7ab 100644
/**
* register_tracer - register a tracer with the ftrace system.
* @type - the plugin for the tracer
@@ -3196,6 +3204,7 @@ static int tracing_release_pipe(struct inode *inode, struct file *file)
@@ -3303,6 +3311,7 @@ static int tracing_release_pipe(struct i
return 0;
}
@ -48,7 +47,7 @@ index ff03353..f4de7ab 100644
static unsigned int
tracing_poll_pipe(struct file *filp, poll_table *poll_table)
{
@@ -3217,8 +3226,7 @@ tracing_poll_pipe(struct file *filp, poll_table *poll_table)
@@ -3324,8 +3333,7 @@ tracing_poll_pipe(struct file *filp, pol
}
}
@ -58,7 +57,7 @@ index ff03353..f4de7ab 100644
{
DEFINE_WAIT(wait);
@@ -3229,6 +3237,20 @@ void default_wait_pipe(struct trace_iterator *iter)
@@ -3336,6 +3344,20 @@ void default_wait_pipe(struct trace_iter
finish_wait(&trace_wait, &wait);
}
@ -79,11 +78,11 @@ index ff03353..f4de7ab 100644
/*
* This is a make-shift waitqueue.
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 092e1f8..69b8700 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -345,7 +345,6 @@ void trace_init_global_iter(struct trace_iterator *iter);
Index: linux-3.4/kernel/trace/trace.h
===================================================================
--- linux-3.4.orig/kernel/trace/trace.h
+++ linux-3.4/kernel/trace/trace.h
@@ -354,7 +354,6 @@ void trace_init_global_iter(struct trace
void tracing_iter_reset(struct trace_iterator *iter, int cpu);
@ -91,6 +90,3 @@ index 092e1f8..69b8700 100644
void poll_wait_pipe(struct trace_iterator *iter);
void ftrace(struct trace_array *tr,
--
1.7.10

48
debian/patches/features/all/rt/0164-ftrace-migrate-disable-tracing.patch.patch → debian/patches/features/all/rt/ftrace-migrate-disable-tracing.patch vendored
View File

@ -1,7 +1,6 @@
From 14e283b0c6f5247b630dda78f0b4bb47c728dd7b Mon Sep 17 00:00:00 2001
Subject: ftrace-migrate-disable-tracing.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 17 Jul 2011 21:56:42 +0200
Subject: [PATCH 164/267] ftrace-migrate-disable-tracing.patch
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
@ -11,10 +10,10 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kernel/trace/trace_output.c | 5 +++++
4 files changed, 14 insertions(+), 4 deletions(-)
diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h
index c3da42d..7c5e176 100644
--- a/include/linux/ftrace_event.h
+++ b/include/linux/ftrace_event.h
Index: linux-3.4/include/linux/ftrace_event.h
===================================================================
--- linux-3.4.orig/include/linux/ftrace_event.h
+++ linux-3.4/include/linux/ftrace_event.h
@@ -49,7 +49,8 @@ struct trace_entry {
unsigned char flags;
unsigned char preempt_count;
@ -25,11 +24,11 @@ index c3da42d..7c5e176 100644
};
#define FTRACE_MAX_EVENT \
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 697e49d..c44456b 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -1123,6 +1123,8 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
Index: linux-3.4/kernel/trace/trace.c
===================================================================
--- linux-3.4.orig/kernel/trace/trace.c
+++ linux-3.4/kernel/trace/trace.c
@@ -1179,6 +1179,8 @@ tracing_generic_entry_update(struct trac
((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) |
((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
(need_resched() ? TRACE_FLAG_NEED_RESCHED : 0);
@ -38,7 +37,7 @@ index 697e49d..c44456b 100644
}
EXPORT_SYMBOL_GPL(tracing_generic_entry_update);
@@ -1854,9 +1856,10 @@ static void print_lat_help_header(struct seq_file *m)
@@ -1937,9 +1939,10 @@ static void print_lat_help_header(struct
seq_puts(m, "# | / _----=> need-resched \n");
seq_puts(m, "# || / _---=> hardirq/softirq \n");
seq_puts(m, "# ||| / _--=> preempt-depth \n");
@ -51,12 +50,12 @@ index 697e49d..c44456b 100644
+ seq_puts(m, "# \\ / ||||| \\ | / \n");
}
static void print_func_help_header(struct seq_file *m)
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index c212a7f..aca63cc 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -116,6 +116,7 @@ static int trace_define_common_fields(void)
static void print_event_info(struct trace_array *tr, struct seq_file *m)
Index: linux-3.4/kernel/trace/trace_events.c
===================================================================
--- linux-3.4.orig/kernel/trace/trace_events.c
+++ linux-3.4/kernel/trace/trace_events.c
@@ -116,6 +116,7 @@ static int trace_define_common_fields(vo
__common_field(unsigned char, flags);
__common_field(unsigned char, preempt_count);
__common_field(int, pid);
@ -64,11 +63,11 @@ index c212a7f..aca63cc 100644
__common_field(int, padding);
return ret;
diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c
index 5199930..c93de84 100644
--- a/kernel/trace/trace_output.c
+++ b/kernel/trace/trace_output.c
@@ -591,6 +591,11 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry)
Index: linux-3.4/kernel/trace/trace_output.c
===================================================================
--- linux-3.4.orig/kernel/trace/trace_output.c
+++ linux-3.4/kernel/trace/trace_output.c
@@ -593,6 +593,11 @@ int trace_print_lat_fmt(struct trace_seq
else
ret = trace_seq_putc(s, '.');
@ -80,6 +79,3 @@ index 5199930..c93de84 100644
return ret;
}
--
1.7.10

35
debian/patches/features/all/rt/0193-futex-Fix-bug-on-when-a-requeued-RT-task-times-out.patch → debian/patches/features/all/rt/futex-requeue-pi-fix.patch vendored
View File

@ -1,12 +1,11 @@
From 3e18d19f9ef2d8e08827a5710c402485b169004c Mon Sep 17 00:00:00 2001
Subject: futex: Fix bug on when a requeued RT task times out
From: Steven Rostedt <rostedt@goodmis.org>
Date: Tue, 10 Apr 2012 14:34:13 -0400
Subject: [PATCH 193/267] futex: Fix bug on when a requeued RT task times out
Requeue with timeout causes a bug with PREEMPT_RT_FULL.
The bug comes from a timed out condition.
TASK 1 TASK 2
------ ------
futex_wait_requeue_pi()
@ -16,12 +15,13 @@ The bug comes from a timed out condition.
double_lock_hb();
raw_spin_lock(pi_lock);
if (current->pi_blocked_on) {
if (current->pi_blocked_on) {
} else {
current->pi_blocked_on = PI_WAKE_INPROGRESS;
run_spin_unlock(pi_lock);
spin_lock(hb->lock); <-- blocked!
plist_for_each_entry_safe(this) {
rt_mutex_start_proxy_lock();
task_blocks_on_rt_mutex();
@ -47,16 +47,18 @@ appropriately.
Cc: stable-rt@vger.kernel.org
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/rtmutex.c | 32 +++++++++++++++++++++++++++++++-
kernel/rtmutex_common.h | 1 +
2 files changed, 32 insertions(+), 1 deletion(-)
diff --git a/kernel/rtmutex.c b/kernel/rtmutex.c
index 13b3c92..f8dcb7b 100644
--- a/kernel/rtmutex.c
+++ b/kernel/rtmutex.c
@@ -69,7 +69,8 @@ static void fixup_rt_mutex_waiters(struct rt_mutex *lock)
Index: linux-3.4/kernel/rtmutex.c
===================================================================
--- linux-3.4.orig/kernel/rtmutex.c
+++ linux-3.4/kernel/rtmutex.c
@@ -69,7 +69,8 @@ static void fixup_rt_mutex_waiters(struc
static int rt_mutex_real_waiter(struct rt_mutex_waiter *waiter)
{
@ -66,7 +68,7 @@ index 13b3c92..f8dcb7b 100644
}
/*
@@ -1010,6 +1011,35 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock,
@@ -1002,6 +1003,35 @@ int rt_mutex_start_proxy_lock(struct rt_
return 1;
}
@ -102,11 +104,11 @@ index 13b3c92..f8dcb7b 100644
ret = task_blocks_on_rt_mutex(lock, waiter, task, detect_deadlock);
if (ret && !rt_mutex_owner(lock)) {
diff --git a/kernel/rtmutex_common.h b/kernel/rtmutex_common.h
index b43d832..47290ec 100644
--- a/kernel/rtmutex_common.h
+++ b/kernel/rtmutex_common.h
@@ -104,6 +104,7 @@ static inline struct task_struct *rt_mutex_owner(struct rt_mutex *lock)
Index: linux-3.4/kernel/rtmutex_common.h
===================================================================
--- linux-3.4.orig/kernel/rtmutex_common.h
+++ linux-3.4/kernel/rtmutex_common.h
@@ -104,6 +104,7 @@ static inline struct task_struct *rt_mut
* PI-futex support (proxy locking functions, etc.):
*/
#define PI_WAKEUP_INPROGRESS ((struct rt_mutex_waiter *) 1)
@ -114,6 +116,3 @@ index b43d832..47290ec 100644
extern struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock);
extern void rt_mutex_init_proxy_locked(struct rt_mutex *lock,
--
1.7.10

22
debian/patches/features/all/rt/0036-generic-Use-raw-local-irq-variant-for-generic-cmpxch.patch → debian/patches/features/all/rt/generic-cmpxchg-use-raw-local-irq.patch vendored
View File

@ -1,22 +1,21 @@
From f105d10b4701b0b44c781bc0c37379f3468521a3 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:30 -0500
Subject: [PATCH 036/267] generic: Use raw local irq variant for generic
cmpxchg
Subject: generic: Use raw local irq variant for generic cmpxchg
No point in tracing those.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
include/asm-generic/cmpxchg-local.h | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/include/asm-generic/cmpxchg-local.h b/include/asm-generic/cmpxchg-local.h
index 2533fdd..d8d4c89 100644
--- a/include/asm-generic/cmpxchg-local.h
+++ b/include/asm-generic/cmpxchg-local.h
@@ -21,7 +21,7 @@ static inline unsigned long __cmpxchg_local_generic(volatile void *ptr,
Index: linux-3.4/include/asm-generic/cmpxchg-local.h
===================================================================
--- linux-3.4.orig/include/asm-generic/cmpxchg-local.h
+++ linux-3.4/include/asm-generic/cmpxchg-local.h
@@ -21,7 +21,7 @@ static inline unsigned long __cmpxchg_lo
if (size == 8 && sizeof(unsigned long) != 8)
wrong_size_cmpxchg(ptr);
@ -25,7 +24,7 @@ index 2533fdd..d8d4c89 100644
switch (size) {
case 1: prev = *(u8 *)ptr;
if (prev == old)
@@ -42,7 +42,7 @@ static inline unsigned long __cmpxchg_local_generic(volatile void *ptr,
@@ -42,7 +42,7 @@ static inline unsigned long __cmpxchg_lo
default:
wrong_size_cmpxchg(ptr);
}
@ -34,7 +33,7 @@ index 2533fdd..d8d4c89 100644
return prev;
}
@@ -55,11 +55,11 @@ static inline u64 __cmpxchg64_local_generic(volatile void *ptr,
@@ -55,11 +55,11 @@ static inline u64 __cmpxchg64_local_gene
u64 prev;
unsigned long flags;
@ -48,6 +47,3 @@ index 2533fdd..d8d4c89 100644
return prev;
}
--
1.7.10

70
debian/patches/features/all/rt/genirq-add-default-mask-cmdline-option.patch vendored Normal file
View File

@ -0,0 +1,70 @@
Subject: genirq: Add default affinity mask command line option
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 25 May 2012 16:59:47 +0200
If we isolate CPUs, then we don't want random device interrupts on
them. Even w/o the user space irq balancer enabled we can end up with
irqs on non boot cpus.
Allow to restrict the default irq affinity mask.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
Documentation/kernel-parameters.txt | 9 +++++++++
kernel/irq/irqdesc.c | 21 +++++++++++++++++++--
2 files changed, 28 insertions(+), 2 deletions(-)
Index: linux-3.4/Documentation/kernel-parameters.txt
===================================================================
--- linux-3.4.orig/Documentation/kernel-parameters.txt
+++ linux-3.4/Documentation/kernel-parameters.txt
@@ -1126,6 +1126,15 @@ bytes respectively. Such letter suffixes
See comment before ip2_setup() in
drivers/char/ip2/ip2base.c.
+ irqaffinity= [SMP] Set the default irq affinity mask
+ Format:
+ <cpu number>,...,<cpu number>
+ or
+ <cpu number>-<cpu number>
+ (must be a positive range in ascending order)
+ or a mixture
+ <cpu number>,...,<cpu number>-<cpu number>
+
irqfixup [HW]
When an interrupt is not handled search all handlers
for it. Intended to get systems with badly broken
Index: linux-3.4/kernel/irq/irqdesc.c
===================================================================
--- linux-3.4.orig/kernel/irq/irqdesc.c
+++ linux-3.4/kernel/irq/irqdesc.c
@@ -23,10 +23,27 @@
static struct lock_class_key irq_desc_lock_class;
#if defined(CONFIG_SMP)
+static int __init irq_affinity_setup(char *str)
+{
+ zalloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT);
+ cpulist_parse(str, irq_default_affinity);
+ /*
+ * Set at least the boot cpu. We don't want to end up with
+ * bugreports caused by random comandline masks
+ */
+ cpumask_set_cpu(smp_processor_id(), irq_default_affinity);
+ return 1;
+}
+__setup("irqaffinity=", irq_affinity_setup);
+
static void __init init_irq_default_affinity(void)
{
- alloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT);
- cpumask_setall(irq_default_affinity);
+#ifdef CONFIG_CPUMASK_OFFSTACK
+ if (!irq_default_affinity)
+ zalloc_cpumask_var(&irq_default_affinity, GFP_NOWAIT);
+#endif
+ if (cpumask_empty(irq_default_affinity))
+ cpumask_setall(irq_default_affinity);
}
#else
static void __init init_irq_default_affinity(void)

17
debian/patches/features/all/rt/0097-genirq-disable-irqpoll-on-rt.patch → debian/patches/features/all/rt/genirq-disable-irqpoll-on-rt.patch vendored
View File

@ -1,21 +1,21 @@
From c52440d3a3faff677bbba7d09e44222de48b6cf2 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 3 Jul 2009 08:29:57 -0500
Subject: [PATCH 097/267] genirq: disable irqpoll on -rt
Subject: genirq: disable irqpoll on -rt
Creates long latencies for no value
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/irq/spurious.c | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index dc813a9..d09e0f5 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -341,6 +341,11 @@ MODULE_PARM_DESC(noirqdebug, "Disable irq lockup detection when true");
Index: linux-3.4/kernel/irq/spurious.c
===================================================================
--- linux-3.4.orig/kernel/irq/spurious.c
+++ linux-3.4/kernel/irq/spurious.c
@@ -341,6 +341,11 @@ MODULE_PARM_DESC(noirqdebug, "Disable ir
static int __init irqfixup_setup(char *str)
{
@ -39,6 +39,3 @@ index dc813a9..d09e0f5 100644
irqfixup = 2;
printk(KERN_WARNING "Misrouted IRQ fixup and polling support "
"enabled\n");
--
1.7.10

17
debian/patches/features/all/rt/0096-genirq-Disable-random-call-on-preempt-rt.patch → debian/patches/features/all/rt/genirq-disable-random-call-on-preempt-rt.patch vendored
View File

@ -1,21 +1,21 @@
From f190b6f8506745170cbb3754f21ab36267847483 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Tue, 21 Jul 2009 16:07:37 +0200
Subject: [PATCH 096/267] genirq: Disable random call on preempt-rt
Subject: genirq: Disable random call on preempt-rt
The random call introduces high latencies and is almost
unused. Disable it for -rt.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
kernel/irq/handle.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
index 470d08c..634620c 100644
--- a/kernel/irq/handle.c
+++ b/kernel/irq/handle.c
@@ -156,8 +156,11 @@ handle_irq_event_percpu(struct irq_desc *desc, struct irqaction *action)
Index: linux-3.4/kernel/irq/handle.c
===================================================================
--- linux-3.4.orig/kernel/irq/handle.c
+++ linux-3.4/kernel/irq/handle.c
@@ -172,8 +172,11 @@ handle_irq_event_percpu(struct irq_desc
action = action->next;
} while (action);
@ -27,6 +27,3 @@ index 470d08c..634620c 100644
if (!noirqdebug)
note_interrupt(irq, desc, retval);
--
1.7.10

28
debian/patches/features/all/rt/0098-genirq-force-threading.patch.patch → debian/patches/features/all/rt/genirq-force-threading.patch vendored
View File

@ -1,7 +1,6 @@
From 63d0b60e274fb76c5b8b1a2fd74a5d93eacc656a Mon Sep 17 00:00:00 2001
Subject: genirq-force-threading.patch
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sun, 3 Apr 2011 11:57:29 +0200
Subject: [PATCH 098/267] genirq-force-threading.patch
Date: Sun, 03 Apr 2011 11:57:29 +0200
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
@ -9,11 +8,11 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
kernel/irq/manage.c | 2 ++
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
index ddd6b2a..b9162dc 100644
--- a/include/linux/interrupt.h
+++ b/include/linux/interrupt.h
@@ -396,9 +396,13 @@ static inline int disable_irq_wake(unsigned int irq)
Index: linux-3.4/include/linux/interrupt.h
===================================================================
--- linux-3.4.orig/include/linux/interrupt.h
+++ linux-3.4/include/linux/interrupt.h
@@ -394,9 +394,13 @@ static inline int disable_irq_wake(unsig
#ifdef CONFIG_IRQ_FORCED_THREADING
@ -29,10 +28,10 @@ index ddd6b2a..b9162dc 100644
#endif
#ifndef __ARCH_SET_SOFTIRQ_PENDING
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 7600092..b3e6228 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
Index: linux-3.4/kernel/irq/manage.c
===================================================================
--- linux-3.4.orig/kernel/irq/manage.c
+++ linux-3.4/kernel/irq/manage.c
@@ -18,6 +18,7 @@
#include "internals.h"
@ -41,7 +40,7 @@ index 7600092..b3e6228 100644
__read_mostly bool force_irqthreads;
static int __init setup_forced_irqthreads(char *arg)
@@ -26,6 +27,7 @@ static int __init setup_forced_irqthreads(char *arg)
@@ -26,6 +27,7 @@ static int __init setup_forced_irqthread
return 0;
}
early_param("threadirqs", setup_forced_irqthreads);
@ -49,6 +48,3 @@ index 7600092..b3e6228 100644
#endif
/**
--
1.7.10

16
debian/patches/features/all/rt/0095-genirq-Disable-DEBUG_SHIRQ-for-rt.patch → debian/patches/features/all/rt/genirq-nodebug-shirq.patch vendored
View File

@ -1,17 +1,16 @@
From 5b808390bbd3ef32efd5d44116bd0e5a17860200 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 18 Mar 2011 10:22:04 +0100
Subject: [PATCH 095/267] genirq: Disable DEBUG_SHIRQ for rt
Date: Sun, 18 Mar 2011 10:22:04 +0100
Subject: genirq: Disable DEBUG_SHIRQ for rt
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
lib/Kconfig.debug | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 82928f5..c347db3 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
Index: linux-3.4/lib/Kconfig.debug
===================================================================
--- linux-3.4.orig/lib/Kconfig.debug
+++ linux-3.4/lib/Kconfig.debug
@@ -151,7 +151,7 @@ config DEBUG_KERNEL
config DEBUG_SHIRQ
@ -21,6 +20,3 @@ index 82928f5..c347db3 100644
help
Enable this to generate a spurious interrupt as soon as a shared
interrupt handler is registered, and just before one is deregistered.
--
1.7.10

18
debian/patches/features/all/rt/0183-hardirq.h-Define-softirq_count-as-OUL-to-kill-build-.patch → debian/patches/features/all/rt/harirq-h-define-softirq_count-as-oul-to-kill-build-warning.patch vendored
View File

@ -1,11 +1,6 @@
From ac220f3571639b8a16d5df2a4fd438938d82afc6 Mon Sep 17 00:00:00 2001
Subject: hardirq.h: Define softirq_count() as OUL to kill build warning
From: Yong Zhang <yong.zhang0@gmail.com>
Date: Thu, 13 Oct 2011 17:19:09 +0800
Subject: [PATCH 183/267] hardirq.h: Define softirq_count() as OUL to kill
build warning
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
kernel/lockdep.c: In function ‘print_bad_irq_dependency’:
kernel/lockdep.c:1476:3: warning: format ‘%lu’ expects type ‘long unsigned int’, but argument 7 has type ‘unsigned int’
@ -27,10 +22,10 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
include/linux/hardirq.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h
index 2f5d318..7059ce2 100644
--- a/include/linux/hardirq.h
+++ b/include/linux/hardirq.h
Index: linux-3.4/include/linux/hardirq.h
===================================================================
--- linux-3.4.orig/include/linux/hardirq.h
+++ linux-3.4/include/linux/hardirq.h
@@ -84,7 +84,7 @@
# define softirq_count() (preempt_count() & SOFTIRQ_MASK)
# define in_serving_softirq() (softirq_count() & SOFTIRQ_OFFSET)
@ -40,6 +35,3 @@ index 2f5d318..7059ce2 100644
extern int in_serving_softirq(void);
#endif
--
1.7.10

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