diff options
author | Ben Hutchings <ben@decadent.org.uk> | 2016-01-15 13:23:17 +0000 |
---|---|---|
committer | Ben Hutchings <ben@decadent.org.uk> | 2016-01-15 13:23:17 +0000 |
commit | fbee7ded77e949f81e0591f68fa2f4d41772c2cc (patch) | |
tree | 9171fe53855c1f51dd636e3ecf53a3f6ff4a78f5 | |
parent | 2f5878146dee3b362212e9f513704b35e006b878 (diff) | |
download | kernel_replicant_linux-fbee7ded77e949f81e0591f68fa2f4d41772c2cc.tar.gz kernel_replicant_linux-fbee7ded77e949f81e0591f68fa2f4d41772c2cc.tar.bz2 kernel_replicant_linux-fbee7ded77e949f81e0591f68fa2f4d41772c2cc.zip |
[rt] Update to 4.4-rt2
258 files changed, 319 insertions, 354 deletions
diff --git a/debian/changelog b/debian/changelog index 563e82c4d791..71ca4245a538 100644 --- a/debian/changelog +++ b/debian/changelog @@ -5,6 +5,7 @@ linux (4.4-1~exp1) UNRELEASED; urgency=medium [ Ben Hutchings ] * [armhf] Enable EXTCON_USB_GPIO, ENSORS_GPIO_FAN as modules, and USB_DWC3_DUAL_ROLE instead of USB_DWC3_HOST (Closes: #810048) + * [rt] Update to 4.4-rt2 -- Ben Hutchings <ben@decadent.org.uk> Wed, 06 Jan 2016 01:28:46 +0000 diff --git a/debian/patches/features/all/rt/0005-futex-Ensure-lock-unlock-symetry-versus-pi_lock-and-.patch b/debian/patches/features/all/rt/0005-futex-Ensure-lock-unlock-symetry-versus-pi_lock-and-.patch index a3b187ce8484..6a0038e05cf2 100644 --- a/debian/patches/features/all/rt/0005-futex-Ensure-lock-unlock-symetry-versus-pi_lock-and-.patch +++ b/debian/patches/features/all/rt/0005-futex-Ensure-lock-unlock-symetry-versus-pi_lock-and-.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 1 Mar 2013 11:17:42 +0100 Subject: futex: Ensure lock/unlock symetry versus pi_lock and hash bucket lock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz In exit_pi_state_list() we have the following locking construct: diff --git a/debian/patches/features/all/rt/0009-ARM-OMAP2-Drop-the-concept-of-certain-power-domains-.patch b/debian/patches/features/all/rt/0009-ARM-OMAP2-Drop-the-concept-of-certain-power-domains-.patch index 7492ea5fdf82..7ec2cc6a7a8a 100644 --- a/debian/patches/features/all/rt/0009-ARM-OMAP2-Drop-the-concept-of-certain-power-domains-.patch +++ b/debian/patches/features/all/rt/0009-ARM-OMAP2-Drop-the-concept-of-certain-power-domains-.patch @@ -3,7 +3,7 @@ From: Russ Dill <Russ.Dill@ti.com> Date: Wed, 5 Aug 2015 15:30:44 +0530 Subject: [PATCH 09/21] ARM: OMAP2: Drop the concept of certain power domains not being able to lose context. -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz It isn't much of a win, and with hibernation, everything loses context. diff --git a/debian/patches/features/all/rt/ARM-enable-irq-in-translation-section-permission-fau.patch b/debian/patches/features/all/rt/ARM-enable-irq-in-translation-section-permission-fau.patch index 992e437be358..15a2a2280ddc 100644 --- a/debian/patches/features/all/rt/ARM-enable-irq-in-translation-section-permission-fau.patch +++ b/debian/patches/features/all/rt/ARM-enable-irq-in-translation-section-permission-fau.patch @@ -1,7 +1,7 @@ From: "Yadi.hu" <yadi.hu@windriver.com> Date: Wed, 10 Dec 2014 10:32:09 +0800 Subject: ARM: enable irq in translation/section permission fault handlers -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Probably happens on all ARM, with CONFIG_PREEMPT_RT_FULL diff --git a/debian/patches/features/all/rt/HACK-printk-drop-the-logbuf_lock-more-often.patch b/debian/patches/features/all/rt/HACK-printk-drop-the-logbuf_lock-more-often.patch index e9f45d219b3f..e680b77941dd 100644 --- a/debian/patches/features/all/rt/HACK-printk-drop-the-logbuf_lock-more-often.patch +++ b/debian/patches/features/all/rt/HACK-printk-drop-the-logbuf_lock-more-often.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 21 Mar 2013 19:01:05 +0100 Subject: printk: Drop the logbuf_lock more often -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The lock is hold with irgs off. The latency drops 500us+ on my arm bugs with a "full" buffer after executing "dmesg" on the shell. diff --git a/debian/patches/features/all/rt/KVM-lapic-mark-LAPIC-timer-handler-as-irqsafe.patch b/debian/patches/features/all/rt/KVM-lapic-mark-LAPIC-timer-handler-as-irqsafe.patch index abf61656698d..57c4094e02d5 100644 --- a/debian/patches/features/all/rt/KVM-lapic-mark-LAPIC-timer-handler-as-irqsafe.patch +++ b/debian/patches/features/all/rt/KVM-lapic-mark-LAPIC-timer-handler-as-irqsafe.patch @@ -1,7 +1,7 @@ From: Marcelo Tosatti <mtosatti@redhat.com> Date: Wed, 8 Apr 2015 20:33:25 -0300 Subject: KVM: lapic: mark LAPIC timer handler as irqsafe -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Since lapic timer handler only wakes up a simple waitqueue, it can be executed from hardirq context. diff --git a/debian/patches/features/all/rt/KVM-use-simple-waitqueue-for-vcpu-wq.patch b/debian/patches/features/all/rt/KVM-use-simple-waitqueue-for-vcpu-wq.patch index 47668c93ce55..76d4797ed8ee 100644 --- a/debian/patches/features/all/rt/KVM-use-simple-waitqueue-for-vcpu-wq.patch +++ b/debian/patches/features/all/rt/KVM-use-simple-waitqueue-for-vcpu-wq.patch @@ -1,7 +1,7 @@ From: Marcelo Tosatti <mtosatti@redhat.com> Date: Wed, 8 Apr 2015 20:33:24 -0300 Subject: KVM: use simple waitqueue for vcpu->wq -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The problem: @@ -127,7 +127,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> ++vcpu->stat.halt_wakeup; } -@@ -701,8 +701,8 @@ int kvmppc_pseries_do_hcall(struct kvm_v +@@ -707,8 +707,8 @@ int kvmppc_pseries_do_hcall(struct kvm_v tvcpu->arch.prodded = 1; smp_mb(); if (vcpu->arch.ceded) { @@ -138,7 +138,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> vcpu->stat.halt_wakeup++; } } -@@ -1441,7 +1441,7 @@ static struct kvmppc_vcore *kvmppc_vcore +@@ -1447,7 +1447,7 @@ static struct kvmppc_vcore *kvmppc_vcore INIT_LIST_HEAD(&vcore->runnable_threads); spin_lock_init(&vcore->lock); spin_lock_init(&vcore->stoltb_lock); @@ -147,7 +147,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> vcore->preempt_tb = TB_NIL; vcore->lpcr = kvm->arch.lpcr; vcore->first_vcpuid = core * threads_per_subcore; -@@ -2513,10 +2513,9 @@ static void kvmppc_vcore_blocked(struct +@@ -2519,10 +2519,9 @@ static void kvmppc_vcore_blocked(struct { struct kvm_vcpu *vcpu; int do_sleep = 1; @@ -160,7 +160,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> /* * Check one last time for pending exceptions and ceded state after -@@ -2530,7 +2529,7 @@ static void kvmppc_vcore_blocked(struct +@@ -2536,7 +2535,7 @@ static void kvmppc_vcore_blocked(struct } if (!do_sleep) { @@ -169,7 +169,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> return; } -@@ -2538,7 +2537,7 @@ static void kvmppc_vcore_blocked(struct +@@ -2544,7 +2543,7 @@ static void kvmppc_vcore_blocked(struct trace_kvmppc_vcore_blocked(vc, 0); spin_unlock(&vc->lock); schedule(); @@ -178,7 +178,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> spin_lock(&vc->lock); vc->vcore_state = VCORE_INACTIVE; trace_kvmppc_vcore_blocked(vc, 1); -@@ -2594,7 +2593,7 @@ static int kvmppc_run_vcpu(struct kvm_ru +@@ -2600,7 +2599,7 @@ static int kvmppc_run_vcpu(struct kvm_ru kvmppc_start_thread(vcpu, vc); trace_kvm_guest_enter(vcpu); } else if (vc->vcore_state == VCORE_SLEEPING) { diff --git a/debian/patches/features/all/rt/acpi-rt-Convert-acpi_gbl_hardware-lock-back-to-a-raw.patch b/debian/patches/features/all/rt/acpi-rt-Convert-acpi_gbl_hardware-lock-back-to-a-raw.patch index cfe942010505..5f2f69def5a4 100644 --- a/debian/patches/features/all/rt/acpi-rt-Convert-acpi_gbl_hardware-lock-back-to-a-raw.patch +++ b/debian/patches/features/all/rt/acpi-rt-Convert-acpi_gbl_hardware-lock-back-to-a-raw.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Wed, 13 Feb 2013 09:26:05 -0500 Subject: acpi/rt: Convert acpi_gbl_hardware lock back to a raw_spinlock_t -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz We hit the following bug with 3.6-rt: diff --git a/debian/patches/features/all/rt/arch-arm64-Add-lazy-preempt-support.patch b/debian/patches/features/all/rt/arch-arm64-Add-lazy-preempt-support.patch index e722c934c964..2280f3f3a473 100644 --- a/debian/patches/features/all/rt/arch-arm64-Add-lazy-preempt-support.patch +++ b/debian/patches/features/all/rt/arch-arm64-Add-lazy-preempt-support.patch @@ -1,7 +1,7 @@ From: Anders Roxell <anders.roxell@linaro.org> Date: Thu, 14 May 2015 17:52:17 +0200 Subject: arch/arm64: Add lazy preempt support -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz arm64 is missing support for PREEMPT_RT. The main feature which is lacking is support for lazy preemption. The arch-specific entry code, diff --git a/debian/patches/features/all/rt/arm-at91-pit-remove-irq-handler-when-clock-is-unused.patch b/debian/patches/features/all/rt/arm-at91-pit-remove-irq-handler-when-clock-is-unused.patch index 75a3c6f60448..5ed2756591d5 100644 --- a/debian/patches/features/all/rt/arm-at91-pit-remove-irq-handler-when-clock-is-unused.patch +++ b/debian/patches/features/all/rt/arm-at91-pit-remove-irq-handler-when-clock-is-unused.patch @@ -1,7 +1,7 @@ From: Benedikt Spranger <b.spranger@linutronix.de> Date: Sat, 6 Mar 2010 17:47:10 +0100 Subject: ARM: AT91: PIT: Remove irq handler when clock event is unused -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Setup and remove the interrupt handler in clock event mode selection. This avoids calling the (shared) interrupt handler when the device is diff --git a/debian/patches/features/all/rt/arm-at91-tclib-default-to-tclib-timer-for-rt.patch b/debian/patches/features/all/rt/arm-at91-tclib-default-to-tclib-timer-for-rt.patch index 67d170e33d8c..a29415a0c6ab 100644 --- a/debian/patches/features/all/rt/arm-at91-tclib-default-to-tclib-timer-for-rt.patch +++ b/debian/patches/features/all/rt/arm-at91-tclib-default-to-tclib-timer-for-rt.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sat, 1 May 2010 18:29:35 +0200 Subject: ARM: at91: tclib: Default to tclib timer for RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz RT is not too happy about the shared timer interrupt in AT91 devices. Default to tclib timer for RT. diff --git a/debian/patches/features/all/rt/arm-convert-boot-lock-to-raw.patch b/debian/patches/features/all/rt/arm-convert-boot-lock-to-raw.patch index 0d9818746388..000f874956bb 100644 --- a/debian/patches/features/all/rt/arm-convert-boot-lock-to-raw.patch +++ b/debian/patches/features/all/rt/arm-convert-boot-lock-to-raw.patch @@ -1,7 +1,7 @@ From: Frank Rowand <frank.rowand@am.sony.com> Date: Mon, 19 Sep 2011 14:51:14 -0700 Subject: arm: Convert arm boot_lock to raw -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The arm boot_lock is used by the secondary processor startup code. The locking task is the idle thread, which has idle->sched_class == &idle_sched_class. diff --git a/debian/patches/features/all/rt/arm-enable-highmem-for-rt.patch b/debian/patches/features/all/rt/arm-enable-highmem-for-rt.patch index 17937992bbf0..ba50628428d9 100644 --- a/debian/patches/features/all/rt/arm-enable-highmem-for-rt.patch +++ b/debian/patches/features/all/rt/arm-enable-highmem-for-rt.patch @@ -1,7 +1,7 @@ Subject: arm: Enable highmem for rt From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 13 Feb 2013 11:03:11 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz fixup highmem for ARM. diff --git a/debian/patches/features/all/rt/arm-highmem-flush-tlb-on-unmap.patch b/debian/patches/features/all/rt/arm-highmem-flush-tlb-on-unmap.patch index 0ae16d5b4386..bd9382e9b4d3 100644 --- a/debian/patches/features/all/rt/arm-highmem-flush-tlb-on-unmap.patch +++ b/debian/patches/features/all/rt/arm-highmem-flush-tlb-on-unmap.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Mon, 11 Mar 2013 21:37:27 +0100 Subject: arm/highmem: Flush tlb on unmap -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The tlb should be flushed on unmap and thus make the mapping entry invalid. This is only done in the non-debug case which does not look diff --git a/debian/patches/features/all/rt/arm-preempt-lazy-support.patch b/debian/patches/features/all/rt/arm-preempt-lazy-support.patch index 434fc5400433..df41da34759c 100644 --- a/debian/patches/features/all/rt/arm-preempt-lazy-support.patch +++ b/debian/patches/features/all/rt/arm-preempt-lazy-support.patch @@ -1,7 +1,7 @@ Subject: arm: Add support for lazy preemption From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 31 Oct 2012 12:04:11 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Implement the arm pieces for lazy preempt. diff --git a/debian/patches/features/all/rt/arm-unwind-use_raw_lock.patch b/debian/patches/features/all/rt/arm-unwind-use_raw_lock.patch index 72c9be2a0b19..3d20f77b4c0c 100644 --- a/debian/patches/features/all/rt/arm-unwind-use_raw_lock.patch +++ b/debian/patches/features/all/rt/arm-unwind-use_raw_lock.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 20 Sep 2013 14:31:54 +0200 Subject: arm/unwind: use a raw_spin_lock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Mostly unwind is done with irqs enabled however SLUB may call it with irqs disabled while creating a new SLUB cache. diff --git a/debian/patches/features/all/rt/arm64-xen--Make-XEN-depend-on-non-rt.patch b/debian/patches/features/all/rt/arm64-xen--Make-XEN-depend-on-non-rt.patch index 477a448e5913..572063c09e69 100644 --- a/debian/patches/features/all/rt/arm64-xen--Make-XEN-depend-on-non-rt.patch +++ b/debian/patches/features/all/rt/arm64-xen--Make-XEN-depend-on-non-rt.patch @@ -1,7 +1,7 @@ Subject: arm64/xen: Make XEN depend on !RT From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 12 Oct 2015 11:18:40 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz It's not ready and probably never will be, unless xen folks have a look at it. diff --git a/debian/patches/features/all/rt/ata-disable-interrupts-if-non-rt.patch b/debian/patches/features/all/rt/ata-disable-interrupts-if-non-rt.patch index 8b193f551d38..05be187ce7e9 100644 --- a/debian/patches/features/all/rt/ata-disable-interrupts-if-non-rt.patch +++ b/debian/patches/features/all/rt/ata-disable-interrupts-if-non-rt.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <srostedt@redhat.com> Date: Fri, 3 Jul 2009 08:44:29 -0500 Subject: ata: Do not disable interrupts in ide code for preempt-rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use the local_irq_*_nort variants. diff --git a/debian/patches/features/all/rt/blk-mq-revert-raw-locks-post-pone-notifier-to-POST_D.patchto-POST_D.patch b/debian/patches/features/all/rt/blk-mq-revert-raw-locks-post-pone-notifier-to-POST_D.patchto-POST_D.patch index c3a481b7619c..df87b782685a 100644 --- a/debian/patches/features/all/rt/blk-mq-revert-raw-locks-post-pone-notifier-to-POST_D.patchto-POST_D.patch +++ b/debian/patches/features/all/rt/blk-mq-revert-raw-locks-post-pone-notifier-to-POST_D.patchto-POST_D.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Sat, 3 May 2014 11:00:29 +0200 Subject: blk-mq: revert raw locks, post pone notifier to POST_DEAD -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The blk_mq_cpu_notify_lock should be raw because some CPU down levels are called with interrupts off. The notifier itself calls currently one diff --git a/debian/patches/features/all/rt/block-blk-mq-use-swait.patch b/debian/patches/features/all/rt/block-blk-mq-use-swait.patch index 83d817d2cb4f..2a871b19e940 100644 --- a/debian/patches/features/all/rt/block-blk-mq-use-swait.patch +++ b/debian/patches/features/all/rt/block-blk-mq-use-swait.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 13 Feb 2015 11:01:26 +0100 Subject: block: blk-mq: Use swait -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz | BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:914 | in_atomic(): 1, irqs_disabled(): 0, pid: 255, name: kworker/u257:6 @@ -46,7 +46,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> --- a/block/blk-core.c +++ b/block/blk-core.c -@@ -644,7 +644,7 @@ int blk_queue_enter(struct request_queue +@@ -660,7 +660,7 @@ int blk_queue_enter(struct request_queue if (!gfpflags_allow_blocking(gfp)) return -EBUSY; @@ -55,7 +55,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> !atomic_read(&q->mq_freeze_depth) || blk_queue_dying(q)); if (blk_queue_dying(q)) -@@ -664,7 +664,7 @@ static void blk_queue_usage_counter_rele +@@ -680,7 +680,7 @@ static void blk_queue_usage_counter_rele struct request_queue *q = container_of(ref, struct request_queue, q_usage_counter); @@ -64,7 +64,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> } struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id) -@@ -726,7 +726,7 @@ struct request_queue *blk_alloc_queue_no +@@ -742,7 +742,7 @@ struct request_queue *blk_alloc_queue_no q->bypass_depth = 1; __set_bit(QUEUE_FLAG_BYPASS, &q->queue_flags); diff --git a/debian/patches/features/all/rt/block-mq-don-t-complete-requests-via-IPI.patch b/debian/patches/features/all/rt/block-mq-don-t-complete-requests-via-IPI.patch index 56d13af7340f..a27394cbee3a 100644 --- a/debian/patches/features/all/rt/block-mq-don-t-complete-requests-via-IPI.patch +++ b/debian/patches/features/all/rt/block-mq-don-t-complete-requests-via-IPI.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 29 Jan 2015 15:10:08 +0100 Subject: block/mq: don't complete requests via IPI -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The IPI runs in hardirq context and there are sleeping locks. This patch moves the completion into a workqueue. diff --git a/debian/patches/features/all/rt/block-mq-drop-per-ctx-cpu_lock.patch b/debian/patches/features/all/rt/block-mq-drop-per-ctx-cpu_lock.patch index 0d69ab5d2df6..229f04d6f805 100644 --- a/debian/patches/features/all/rt/block-mq-drop-per-ctx-cpu_lock.patch +++ b/debian/patches/features/all/rt/block-mq-drop-per-ctx-cpu_lock.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed, 18 Feb 2015 18:37:26 +0100 Subject: block/mq: drop per ctx cpu_lock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz While converting the get_cpu() to get_cpu_light() I added a cpu lock to ensure the same code is not invoked twice on the same CPU. And now I run diff --git a/debian/patches/features/all/rt/block-mq-drop-preempt-disable.patch b/debian/patches/features/all/rt/block-mq-drop-preempt-disable.patch index 01deb37d4735..50f92fb63967 100644 --- a/debian/patches/features/all/rt/block-mq-drop-preempt-disable.patch +++ b/debian/patches/features/all/rt/block-mq-drop-preempt-disable.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Tue, 14 Jul 2015 14:26:34 +0200 Subject: block/mq: do not invoke preempt_disable() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz preempt_disable() and get_cpu() don't play well together with the sleeping locks it tries to allocate later. diff --git a/debian/patches/features/all/rt/block-mq-use-cpu_light.patch b/debian/patches/features/all/rt/block-mq-use-cpu_light.patch index e549c68229a9..5ed3de7e6c0d 100644 --- a/debian/patches/features/all/rt/block-mq-use-cpu_light.patch +++ b/debian/patches/features/all/rt/block-mq-use-cpu_light.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed, 9 Apr 2014 10:37:23 +0200 Subject: block: mq: use cpu_light() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz there is a might sleep splat because get_cpu() disables preemption and later we grab a lock. As a workaround for this we use get_cpu_light() diff --git a/debian/patches/features/all/rt/block-shorten-interrupt-disabled-regions.patch b/debian/patches/features/all/rt/block-shorten-interrupt-disabled-regions.patch index af37b663af86..2c4deaf44d2a 100644 --- a/debian/patches/features/all/rt/block-shorten-interrupt-disabled-regions.patch +++ b/debian/patches/features/all/rt/block-shorten-interrupt-disabled-regions.patch @@ -1,7 +1,7 @@ Subject: block: Shorten interrupt disabled regions From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 22 Jun 2011 19:47:02 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Moving the blk_sched_flush_plug() call out of the interrupt/preempt disabled region in the scheduler allows us to replace @@ -48,7 +48,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de --- a/block/blk-core.c +++ b/block/blk-core.c -@@ -3182,7 +3182,7 @@ static void queue_unplugged(struct reque +@@ -3198,7 +3198,7 @@ static void queue_unplugged(struct reque blk_run_queue_async(q); else __blk_run_queue(q); @@ -57,7 +57,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de } static void flush_plug_callbacks(struct blk_plug *plug, bool from_schedule) -@@ -3230,7 +3230,6 @@ EXPORT_SYMBOL(blk_check_plugged); +@@ -3246,7 +3246,6 @@ EXPORT_SYMBOL(blk_check_plugged); void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule) { struct request_queue *q; @@ -65,7 +65,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de struct request *rq; LIST_HEAD(list); unsigned int depth; -@@ -3250,11 +3249,6 @@ void blk_flush_plug_list(struct blk_plug +@@ -3266,11 +3265,6 @@ void blk_flush_plug_list(struct blk_plug q = NULL; depth = 0; @@ -77,7 +77,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de while (!list_empty(&list)) { rq = list_entry_rq(list.next); list_del_init(&rq->queuelist); -@@ -3267,7 +3261,7 @@ void blk_flush_plug_list(struct blk_plug +@@ -3283,7 +3277,7 @@ void blk_flush_plug_list(struct blk_plug queue_unplugged(q, depth, from_schedule); q = rq->q; depth = 0; @@ -86,7 +86,7 @@ Link: http://lkml.kernel.org/r/20110622174919.025446432@linutronix.de } /* -@@ -3294,8 +3288,6 @@ void blk_flush_plug_list(struct blk_plug +@@ -3310,8 +3304,6 @@ void blk_flush_plug_list(struct blk_plug */ if (q) queue_unplugged(q, depth, from_schedule); diff --git a/debian/patches/features/all/rt/block-use-cpu-chill.patch b/debian/patches/features/all/rt/block-use-cpu-chill.patch index e6eff28b3ee1..c44b1d8bea64 100644 --- a/debian/patches/features/all/rt/block-use-cpu-chill.patch +++ b/debian/patches/features/all/rt/block-use-cpu-chill.patch @@ -1,7 +1,7 @@ Subject: block: Use cpu_chill() for retry loops From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 20 Dec 2012 18:28:26 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Retry loops on RT might loop forever when the modifying side was preempted. Steven also observed a live lock when there was a diff --git a/debian/patches/features/all/rt/bug-rt-dependend-variants.patch b/debian/patches/features/all/rt/bug-rt-dependend-variants.patch index 07431c851b01..5652fe93f611 100644 --- a/debian/patches/features/all/rt/bug-rt-dependend-variants.patch +++ b/debian/patches/features/all/rt/bug-rt-dependend-variants.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:58 -0500 Subject: bug: BUG_ON/WARN_ON variants dependend on RT/!RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Introduce RT/NON-RT WARN/BUG statements to avoid ifdefs in the code. diff --git a/debian/patches/features/all/rt/cgroups-scheduling-while-atomic-in-cgroup-code.patch b/debian/patches/features/all/rt/cgroups-scheduling-while-atomic-in-cgroup-code.patch index e2a322120b0d..28c3b8f80d7d 100644 --- a/debian/patches/features/all/rt/cgroups-scheduling-while-atomic-in-cgroup-code.patch +++ b/debian/patches/features/all/rt/cgroups-scheduling-while-atomic-in-cgroup-code.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com> Date: Sat, 21 Jun 2014 10:09:48 +0200 Subject: memcontrol: Prevent scheduling while atomic in cgroup code -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz mm, memcg: make refill_stock() use get_cpu_light() @@ -43,7 +43,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> --- a/mm/memcontrol.c +++ b/mm/memcontrol.c -@@ -1913,14 +1913,17 @@ static void drain_local_stock(struct wor +@@ -1937,14 +1937,17 @@ static void drain_local_stock(struct wor */ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages) { diff --git a/debian/patches/features/all/rt/cgroups-use-simple-wait-in-css_release.patch b/debian/patches/features/all/rt/cgroups-use-simple-wait-in-css_release.patch index cef9cb8f4b0e..a2b6e0d44312 100644 --- a/debian/patches/features/all/rt/cgroups-use-simple-wait-in-css_release.patch +++ b/debian/patches/features/all/rt/cgroups-use-simple-wait-in-css_release.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 13 Feb 2015 15:52:24 +0100 Subject: cgroups: use simple wait in css_release() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz To avoid: |BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:914 diff --git a/debian/patches/features/all/rt/clocksource-tclib-allow-higher-clockrates.patch b/debian/patches/features/all/rt/clocksource-tclib-allow-higher-clockrates.patch index 2ec29a43fd48..31708b68141d 100644 --- a/debian/patches/features/all/rt/clocksource-tclib-allow-higher-clockrates.patch +++ b/debian/patches/features/all/rt/clocksource-tclib-allow-higher-clockrates.patch @@ -1,7 +1,7 @@ From: Benedikt Spranger <b.spranger@linutronix.de> Date: Mon, 8 Mar 2010 18:57:04 +0100 Subject: clocksource: TCLIB: Allow higher clock rates for clock events -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz As default the TCLIB uses the 32KiHz base clock rate for clock events. Add a compile time selection to allow higher clock resulution. diff --git a/debian/patches/features/all/rt/completion-use-simple-wait-queues.patch b/debian/patches/features/all/rt/completion-use-simple-wait-queues.patch index 447fd88a9386..94f7f4c99016 100644 --- a/debian/patches/features/all/rt/completion-use-simple-wait-queues.patch +++ b/debian/patches/features/all/rt/completion-use-simple-wait-queues.patch @@ -1,7 +1,7 @@ Subject: completion: Use simple wait queues From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 11 Jan 2013 11:23:51 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Completions have no long lasting callbacks and therefor do not need the complex waitqueue variant. Use simple waitqueues which reduces the diff --git a/debian/patches/features/all/rt/cond-resched-lock-rt-tweak.patch b/debian/patches/features/all/rt/cond-resched-lock-rt-tweak.patch index 08139960222d..76bf97051dcd 100644 --- a/debian/patches/features/all/rt/cond-resched-lock-rt-tweak.patch +++ b/debian/patches/features/all/rt/cond-resched-lock-rt-tweak.patch @@ -1,7 +1,7 @@ Subject: sched: Use the proper LOCK_OFFSET for cond_resched() From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 22:51:33 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz RT does not increment preempt count when a 'sleeping' spinlock is locked. Update PREEMPT_LOCK_OFFSET for that case. diff --git a/debian/patches/features/all/rt/cond-resched-softirq-rt.patch b/debian/patches/features/all/rt/cond-resched-softirq-rt.patch index 86df56cfb9de..a30f1b6d6099 100644 --- a/debian/patches/features/all/rt/cond-resched-softirq-rt.patch +++ b/debian/patches/features/all/rt/cond-resched-softirq-rt.patch @@ -1,7 +1,7 @@ Subject: sched: Take RT softirq semantics into account in cond_resched() From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 14 Jul 2011 09:56:44 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The softirq semantics work different on -RT. There is no SOFTIRQ_MASK in the preemption counter which leads to the BUG_ON() statement in @@ -16,7 +16,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -2984,12 +2984,16 @@ extern int __cond_resched_lock(spinlock_ +@@ -2986,12 +2986,16 @@ extern int __cond_resched_lock(spinlock_ __cond_resched_lock(lock); \ }) diff --git a/debian/patches/features/all/rt/cpu-hotplug-Document-why-PREEMPT_RT-uses-a-spinlock.patch b/debian/patches/features/all/rt/cpu-hotplug-Document-why-PREEMPT_RT-uses-a-spinlock.patch index 53b1077d92ed..2130d6ce8993 100644 --- a/debian/patches/features/all/rt/cpu-hotplug-Document-why-PREEMPT_RT-uses-a-spinlock.patch +++ b/debian/patches/features/all/rt/cpu-hotplug-Document-why-PREEMPT_RT-uses-a-spinlock.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Thu, 5 Dec 2013 09:16:52 -0500 Subject: cpu hotplug: Document why PREEMPT_RT uses a spinlock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The patch: diff --git a/debian/patches/features/all/rt/cpu-rt-make-hotplug-lock-a-sleeping-spinlock-on-rt.patch b/debian/patches/features/all/rt/cpu-rt-make-hotplug-lock-a-sleeping-spinlock-on-rt.patch index 77f9701e9b81..47e18b6dce01 100644 --- a/debian/patches/features/all/rt/cpu-rt-make-hotplug-lock-a-sleeping-spinlock-on-rt.patch +++ b/debian/patches/features/all/rt/cpu-rt-make-hotplug-lock-a-sleeping-spinlock-on-rt.patch @@ -1,7 +1,7 @@ Subject: cpu: Make hotplug.lock a "sleeping" spinlock on RT From: Steven Rostedt <rostedt@goodmis.org> Date: Fri, 02 Mar 2012 10:36:57 -0500 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Tasks can block on hotplug.lock in pin_current_cpu(), but their state might be != RUNNING. So the mutex wakeup will set the state diff --git a/debian/patches/features/all/rt/cpu-rt-rework-cpu-down.patch b/debian/patches/features/all/rt/cpu-rt-rework-cpu-down.patch index 90d43d5f0080..fdb5ab7a0a82 100644 --- a/debian/patches/features/all/rt/cpu-rt-rework-cpu-down.patch +++ b/debian/patches/features/all/rt/cpu-rt-rework-cpu-down.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <srostedt@redhat.com> Date: Mon, 16 Jul 2012 08:07:43 +0000 Subject: cpu/rt: Rework cpu down for PREEMPT_RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Bringing a CPU down is a pain with the PREEMPT_RT kernel because tasks can be preempted in many more places than in non-RT. In @@ -57,7 +57,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -2282,6 +2282,10 @@ extern void do_set_cpus_allowed(struct t +@@ -2284,6 +2284,10 @@ extern void do_set_cpus_allowed(struct t extern int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask); @@ -68,7 +68,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> #else static inline void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) -@@ -2294,6 +2298,9 @@ static inline int set_cpus_allowed_ptr(s +@@ -2296,6 +2300,9 @@ static inline int set_cpus_allowed_ptr(s return -EINVAL; return 0; } diff --git a/debian/patches/features/all/rt/cpu_chill-Add-a-UNINTERRUPTIBLE-hrtimer_nanosleep.patch b/debian/patches/features/all/rt/cpu_chill-Add-a-UNINTERRUPTIBLE-hrtimer_nanosleep.patch index 457870e510f2..de048d9d86ac 100644 --- a/debian/patches/features/all/rt/cpu_chill-Add-a-UNINTERRUPTIBLE-hrtimer_nanosleep.patch +++ b/debian/patches/features/all/rt/cpu_chill-Add-a-UNINTERRUPTIBLE-hrtimer_nanosleep.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Tue, 4 Mar 2014 12:28:32 -0500 Subject: cpu_chill: Add a UNINTERRUPTIBLE hrtimer_nanosleep -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz We hit another bug that was caused by switching cpu_chill() from msleep() to hrtimer_nanosleep(). diff --git a/debian/patches/features/all/rt/cpu_down_move_migrate_enable_back.patch b/debian/patches/features/all/rt/cpu_down_move_migrate_enable_back.patch index c6d02be1c064..3ac0941b7dfe 100644 --- a/debian/patches/features/all/rt/cpu_down_move_migrate_enable_back.patch +++ b/debian/patches/features/all/rt/cpu_down_move_migrate_enable_back.patch @@ -1,7 +1,7 @@ From: Tiejun Chen <tiejun.chen@windriver.com> Subject: cpu_down: move migrate_enable() back Date: Thu, 7 Nov 2013 10:06:07 +0800 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Commit 08c1ab68, "hotplug-use-migrate-disable.patch", intends to use migrate_enable()/migrate_disable() to replace that combination diff --git a/debian/patches/features/all/rt/cpufreq-drop-K8-s-driver-from-beeing-selected.patch b/debian/patches/features/all/rt/cpufreq-drop-K8-s-driver-from-beeing-selected.patch index bec490b78b80..3080ec7b59d0 100644 --- a/debian/patches/features/all/rt/cpufreq-drop-K8-s-driver-from-beeing-selected.patch +++ b/debian/patches/features/all/rt/cpufreq-drop-K8-s-driver-from-beeing-selected.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 9 Apr 2015 15:23:01 +0200 Subject: cpufreq: drop K8's driver from beeing selected -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Ralf posted a picture of a backtrace from diff --git a/debian/patches/features/all/rt/cpumask-disable-offstack-on-rt.patch b/debian/patches/features/all/rt/cpumask-disable-offstack-on-rt.patch index 7132934201d9..e7e0e4070776 100644 --- a/debian/patches/features/all/rt/cpumask-disable-offstack-on-rt.patch +++ b/debian/patches/features/all/rt/cpumask-disable-offstack-on-rt.patch @@ -1,7 +1,7 @@ Subject: cpumask: Disable CONFIG_CPUMASK_OFFSTACK for RT From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 14 Dec 2011 01:03:49 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz 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. diff --git a/debian/patches/features/all/rt/crypto-Reduce-preempt-disabled-regions-more-algos.patch b/debian/patches/features/all/rt/crypto-Reduce-preempt-disabled-regions-more-algos.patch index 520a3e48380b..7ed97af4a2ba 100644 --- a/debian/patches/features/all/rt/crypto-Reduce-preempt-disabled-regions-more-algos.patch +++ b/debian/patches/features/all/rt/crypto-Reduce-preempt-disabled-regions-more-algos.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 21 Feb 2014 17:24:04 +0100 Subject: crypto: Reduce preempt disabled regions, more algos -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Don Estabrook reported | kernel: WARNING: CPU: 2 PID: 858 at kernel/sched/core.c:2428 migrate_disable+0xed/0x100() diff --git a/debian/patches/features/all/rt/debugobjects-rt.patch b/debian/patches/features/all/rt/debugobjects-rt.patch index 60363e66aa12..c4aebf429029 100644 --- a/debian/patches/features/all/rt/debugobjects-rt.patch +++ b/debian/patches/features/all/rt/debugobjects-rt.patch @@ -1,7 +1,7 @@ Subject: debugobjects: Make RT aware From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 21:41:35 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Avoid filling the pool / allocating memory with irqs off(). diff --git a/debian/patches/features/all/rt/dm-make-rt-aware.patch b/debian/patches/features/all/rt/dm-make-rt-aware.patch index c632a30b0e6f..fb94382b1fcd 100644 --- a/debian/patches/features/all/rt/dm-make-rt-aware.patch +++ b/debian/patches/features/all/rt/dm-make-rt-aware.patch @@ -1,7 +1,7 @@ Subject: dm: Make rt aware From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 14 Nov 2011 23:06:09 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use the BUG_ON_NORT variant for the irq_disabled() checks. RT has interrupts legitimately enabled here as we cant deadlock against the diff --git a/debian/patches/features/all/rt/drivers-net-8139-disable-irq-nosync.patch b/debian/patches/features/all/rt/drivers-net-8139-disable-irq-nosync.patch index b060d994298a..505e55472a99 100644 --- a/debian/patches/features/all/rt/drivers-net-8139-disable-irq-nosync.patch +++ b/debian/patches/features/all/rt/drivers-net-8139-disable-irq-nosync.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:24 -0500 Subject: drivers/net: Use disable_irq_nosync() in 8139too -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use disable_irq_nosync() instead of disable_irq() as this might be called in atomic context with netpoll. diff --git a/debian/patches/features/all/rt/drivers-net-fix-livelock-issues.patch b/debian/patches/features/all/rt/drivers-net-fix-livelock-issues.patch index bfd111fc8fe8..edd3be824271 100644 --- a/debian/patches/features/all/rt/drivers-net-fix-livelock-issues.patch +++ b/debian/patches/features/all/rt/drivers-net-fix-livelock-issues.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sat, 20 Jun 2009 11:36:54 +0200 Subject: drivers/net: fix livelock issues -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Preempt-RT runs into a live lock issue with the NETDEV_TX_LOCKED micro optimization. The reason is that the softirq thread is rescheduling diff --git a/debian/patches/features/all/rt/drivers-net-vortex-fix-locking-issues.patch b/debian/patches/features/all/rt/drivers-net-vortex-fix-locking-issues.patch index ddd2861700c5..cf1337346d0c 100644 --- a/debian/patches/features/all/rt/drivers-net-vortex-fix-locking-issues.patch +++ b/debian/patches/features/all/rt/drivers-net-vortex-fix-locking-issues.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Fri, 3 Jul 2009 08:30:00 -0500 Subject: drivers/net: vortex fix locking issues -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Argh, cut and paste wasn't enough... diff --git a/debian/patches/features/all/rt/drivers-random-reduce-preempt-disabled-region.patch b/debian/patches/features/all/rt/drivers-random-reduce-preempt-disabled-region.patch index 49a8ad8331a4..7b48d673a3db 100644 --- a/debian/patches/features/all/rt/drivers-random-reduce-preempt-disabled-region.patch +++ b/debian/patches/features/all/rt/drivers-random-reduce-preempt-disabled-region.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:30 -0500 Subject: drivers: random: Reduce preempt disabled region -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz No need to keep preemption disabled across the whole function. diff --git a/debian/patches/features/all/rt/drivers-tty-fix-omap-lock-crap.patch b/debian/patches/features/all/rt/drivers-tty-fix-omap-lock-crap.patch index a841023e07e4..7a5c117b5f0a 100644 --- a/debian/patches/features/all/rt/drivers-tty-fix-omap-lock-crap.patch +++ b/debian/patches/features/all/rt/drivers-tty-fix-omap-lock-crap.patch @@ -1,7 +1,7 @@ Subject: tty/serial/omap: Make the locking RT aware From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 28 Jul 2011 13:32:57 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The lock is a sleeping lock and local_irq_save() is not the optimsation we are looking for. Redo it to make it work on -RT and diff --git a/debian/patches/features/all/rt/drivers-tty-pl011-irq-disable-madness.patch b/debian/patches/features/all/rt/drivers-tty-pl011-irq-disable-madness.patch index 11224c02fdc5..1ee262aec9e4 100644 --- a/debian/patches/features/all/rt/drivers-tty-pl011-irq-disable-madness.patch +++ b/debian/patches/features/all/rt/drivers-tty-pl011-irq-disable-madness.patch @@ -1,7 +1,7 @@ Subject: tty/serial/pl011: Make the locking work on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 08 Jan 2013 21:36:51 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The lock is a sleeping lock and local_irq_save() is not the optimsation we are looking for. Redo it to make it work on -RT and non-RT. diff --git a/debian/patches/features/all/rt/drm-i915-drop-trace_i915_gem_ring_dispatch-onrt.patch b/debian/patches/features/all/rt/drm-i915-drop-trace_i915_gem_ring_dispatch-onrt.patch index 9fc80a220cef..53b5796e14f7 100644 --- a/debian/patches/features/all/rt/drm-i915-drop-trace_i915_gem_ring_dispatch-onrt.patch +++ b/debian/patches/features/all/rt/drm-i915-drop-trace_i915_gem_ring_dispatch-onrt.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 25 Apr 2013 18:12:52 +0200 Subject: drm/i915: drop trace_i915_gem_ring_dispatch on rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz This tracepoint is responsible for: diff --git a/debian/patches/features/all/rt/dump-stack-don-t-disable-preemption-during-trace.patch b/debian/patches/features/all/rt/dump-stack-don-t-disable-preemption-during-trace.patch index 4ef90636c12a..6f48a7d8e663 100644 --- a/debian/patches/features/all/rt/dump-stack-don-t-disable-preemption-during-trace.patch +++ b/debian/patches/features/all/rt/dump-stack-don-t-disable-preemption-during-trace.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Sun, 16 Aug 2015 14:27:50 +0200 Subject: dump stack: don't disable preemption during trace -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz I see here large latencies during a stack dump on x86. The preempt_disable() and get_cpu() should forbid moving the task to another diff --git a/debian/patches/features/all/rt/epoll-use-get-cpu-light.patch b/debian/patches/features/all/rt/epoll-use-get-cpu-light.patch index d397792b9c06..91a362863b25 100644 --- a/debian/patches/features/all/rt/epoll-use-get-cpu-light.patch +++ b/debian/patches/features/all/rt/epoll-use-get-cpu-light.patch @@ -1,7 +1,7 @@ Subject: fs/epoll: Do not disable preemption on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 08 Jul 2011 16:35:35 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz ep_call_nested() takes a sleeping lock so we can't disable preemption. The light version is enough since ep_call_nested() doesn't mind beeing diff --git a/debian/patches/features/all/rt/fs-aio-simple-simple-work.patch b/debian/patches/features/all/rt/fs-aio-simple-simple-work.patch index eede1009c262..8f9160abd8b3 100644 --- a/debian/patches/features/all/rt/fs-aio-simple-simple-work.patch +++ b/debian/patches/features/all/rt/fs-aio-simple-simple-work.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Mon, 16 Feb 2015 18:49:10 +0100 Subject: fs/aio: simple simple work -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz |BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:768 |in_atomic(): 1, irqs_disabled(): 0, pid: 26, name: rcuos/2 diff --git a/debian/patches/features/all/rt/fs-block-rt-support.patch b/debian/patches/features/all/rt/fs-block-rt-support.patch index 3a993723bf4d..15ca03c9b12e 100644 --- a/debian/patches/features/all/rt/fs-block-rt-support.patch +++ b/debian/patches/features/all/rt/fs-block-rt-support.patch @@ -1,7 +1,7 @@ Subject: block: Turn off warning which is bogus on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 14 Jun 2011 17:05:09 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On -RT the context is always with IRQs enabled. Ignore this warning on -RT. @@ -12,7 +12,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- a/block/blk-core.c +++ b/block/blk-core.c -@@ -217,7 +217,7 @@ EXPORT_SYMBOL(blk_delay_queue); +@@ -233,7 +233,7 @@ EXPORT_SYMBOL(blk_start_queue_async); **/ void blk_start_queue(struct request_queue *q) { diff --git a/debian/patches/features/all/rt/fs-dcache-use-cpu-chill-in-trylock-loops.patch b/debian/patches/features/all/rt/fs-dcache-use-cpu-chill-in-trylock-loops.patch index 77eb9f95f71a..0b3033ef8e94 100644 --- a/debian/patches/features/all/rt/fs-dcache-use-cpu-chill-in-trylock-loops.patch +++ b/debian/patches/features/all/rt/fs-dcache-use-cpu-chill-in-trylock-loops.patch @@ -1,7 +1,7 @@ Subject: fs: dcache: Use cpu_chill() in trylock loops From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 07 Mar 2012 21:00:34 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Retry loops on RT might loop forever when the modifying side was preempted. Use cpu_chill() instead of cpu_relax() to let the system diff --git a/debian/patches/features/all/rt/fs-jbd-replace-bh_state-lock.patch b/debian/patches/features/all/rt/fs-jbd-replace-bh_state-lock.patch index efec4b6fa623..16190fe07545 100644 --- a/debian/patches/features/all/rt/fs-jbd-replace-bh_state-lock.patch +++ b/debian/patches/features/all/rt/fs-jbd-replace-bh_state-lock.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 18 Mar 2011 10:11:25 +0100 Subject: fs: jbd/jbd2: Make state lock and journal head lock rt safe -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz bit_spin_locks break under RT. diff --git a/debian/patches/features/all/rt/fs-jbd2-pull-your-plug-when-waiting-for-space.patch b/debian/patches/features/all/rt/fs-jbd2-pull-your-plug-when-waiting-for-space.patch index fe9796074396..2325cc4b2cb3 100644 --- a/debian/patches/features/all/rt/fs-jbd2-pull-your-plug-when-waiting-for-space.patch +++ b/debian/patches/features/all/rt/fs-jbd2-pull-your-plug-when-waiting-for-space.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Mon, 17 Feb 2014 17:30:03 +0100 Subject: fs: jbd2: pull your plug when waiting for space -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Two cps in parallel managed to stall the the ext4 fs. It seems that journal code is either waiting for locks or sleeping waiting for diff --git a/debian/patches/features/all/rt/fs-namespace-preemption-fix.patch b/debian/patches/features/all/rt/fs-namespace-preemption-fix.patch index c51ee1de31ee..70dfa5769508 100644 --- a/debian/patches/features/all/rt/fs-namespace-preemption-fix.patch +++ b/debian/patches/features/all/rt/fs-namespace-preemption-fix.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 19 Jul 2009 08:44:27 -0500 Subject: fs: namespace preemption fix -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On RT we cannot loop with preemption disabled here as mnt_make_readonly() might have been preempted. We can safely enable diff --git a/debian/patches/features/all/rt/fs-ntfs-disable-interrupt-non-rt.patch b/debian/patches/features/all/rt/fs-ntfs-disable-interrupt-non-rt.patch index f7c72bb344c8..b15b50bfe52f 100644 --- a/debian/patches/features/all/rt/fs-ntfs-disable-interrupt-non-rt.patch +++ b/debian/patches/features/all/rt/fs-ntfs-disable-interrupt-non-rt.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <efault@gmx.de> Date: Fri, 3 Jul 2009 08:44:12 -0500 Subject: fs: ntfs: disable interrupt only on !RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On Sat, 2007-10-27 at 11:44 +0200, Ingo Molnar wrote: > * Nick Piggin <nickpiggin@yahoo.com.au> wrote: diff --git a/debian/patches/features/all/rt/fs-replace-bh_uptodate_lock-for-rt.patch b/debian/patches/features/all/rt/fs-replace-bh_uptodate_lock-for-rt.patch index c9bd3b6826ed..d2746226aca4 100644 --- a/debian/patches/features/all/rt/fs-replace-bh_uptodate_lock-for-rt.patch +++ b/debian/patches/features/all/rt/fs-replace-bh_uptodate_lock-for-rt.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 18 Mar 2011 09:18:52 +0100 Subject: buffer_head: Replace bh_uptodate_lock for -rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Wrap the bit_spin_lock calls into a separate inline and add the RT replacements with a real spinlock. diff --git a/debian/patches/features/all/rt/ftrace-migrate-disable-tracing.patch b/debian/patches/features/all/rt/ftrace-migrate-disable-tracing.patch index 1d9cd66aaa00..1c3297103abc 100644 --- a/debian/patches/features/all/rt/ftrace-migrate-disable-tracing.patch +++ b/debian/patches/features/all/rt/ftrace-migrate-disable-tracing.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 21:56:42 +0200 Subject: trace: Add migrate-disabled counter to tracing output -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- diff --git a/debian/patches/features/all/rt/futex-requeue-pi-fix.patch b/debian/patches/features/all/rt/futex-requeue-pi-fix.patch index 377920ccf13f..1290b0214287 100644 --- a/debian/patches/features/all/rt/futex-requeue-pi-fix.patch +++ b/debian/patches/features/all/rt/futex-requeue-pi-fix.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Tue, 14 Jul 2015 14:26:34 +0200 Subject: futex: Fix bug on when a requeued RT task times out -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Requeue with timeout causes a bug with PREEMPT_RT_FULL. diff --git a/debian/patches/features/all/rt/genirq-disable-irqpoll-on-rt.patch b/debian/patches/features/all/rt/genirq-disable-irqpoll-on-rt.patch index 9741895e1921..99d3b65586ed 100644 --- a/debian/patches/features/all/rt/genirq-disable-irqpoll-on-rt.patch +++ b/debian/patches/features/all/rt/genirq-disable-irqpoll-on-rt.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:57 -0500 Subject: genirq: Disable irqpoll on -rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Creates long latencies for no value diff --git a/debian/patches/features/all/rt/genirq-do-not-invoke-the-affinity-callback-via-a-wor.patch b/debian/patches/features/all/rt/genirq-do-not-invoke-the-affinity-callback-via-a-wor.patch index 65d89ccff365..0c497a36cc60 100644 --- a/debian/patches/features/all/rt/genirq-do-not-invoke-the-affinity-callback-via-a-wor.patch +++ b/debian/patches/features/all/rt/genirq-do-not-invoke-the-affinity-callback-via-a-wor.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed, 21 Aug 2013 17:48:46 +0200 Subject: genirq: Do not invoke the affinity callback via a workqueue on RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Joe Korty reported, that __irq_set_affinity_locked() schedules a workqueue while holding a rawlock which results in a might_sleep() diff --git a/debian/patches/features/all/rt/genirq-force-threading.patch b/debian/patches/features/all/rt/genirq-force-threading.patch index 3931b1f7cd66..d9bfb72f5618 100644 --- a/debian/patches/features/all/rt/genirq-force-threading.patch +++ b/debian/patches/features/all/rt/genirq-force-threading.patch @@ -1,7 +1,7 @@ Subject: genirq: Force interrupt thread on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 03 Apr 2011 11:57:29 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Force threaded_irqs and optimize the code (force_irqthreads) in regard to this. diff --git a/debian/patches/features/all/rt/hotplug-Use-set_cpus_allowed_ptr-in-sync_unplug_thre.patch b/debian/patches/features/all/rt/hotplug-Use-set_cpus_allowed_ptr-in-sync_unplug_thre.patch index d9c3115faa9f..4fbb5e360d0e 100644 --- a/debian/patches/features/all/rt/hotplug-Use-set_cpus_allowed_ptr-in-sync_unplug_thre.patch +++ b/debian/patches/features/all/rt/hotplug-Use-set_cpus_allowed_ptr-in-sync_unplug_thre.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com> Date: Tue, 24 Mar 2015 08:14:49 +0100 Subject: hotplug: Use set_cpus_allowed_ptr() in sync_unplug_thread() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz do_set_cpus_allowed() is not safe vs ->sched_class change. diff --git a/debian/patches/features/all/rt/hotplug-light-get-online-cpus.patch b/debian/patches/features/all/rt/hotplug-light-get-online-cpus.patch index 64b1d5b3afd0..2c5a7a04e906 100644 --- a/debian/patches/features/all/rt/hotplug-light-get-online-cpus.patch +++ b/debian/patches/features/all/rt/hotplug-light-get-online-cpus.patch @@ -1,7 +1,7 @@ Subject: hotplug: Lightweight get online cpus From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 15 Jun 2011 12:36:06 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz get_online_cpus() is a heavy weight function which involves a global mutex. migrate_disable() wants a simpler construct which prevents only diff --git a/debian/patches/features/all/rt/hotplug-sync_unplug-no-27-5cn-27-in-task-name.patch b/debian/patches/features/all/rt/hotplug-sync_unplug-no-27-5cn-27-in-task-name.patch index 06e6d2ddfb22..797f5fe0a7f4 100644 --- a/debian/patches/features/all/rt/hotplug-sync_unplug-no-27-5cn-27-in-task-name.patch +++ b/debian/patches/features/all/rt/hotplug-sync_unplug-no-27-5cn-27-in-task-name.patch @@ -1,7 +1,7 @@ Subject: hotplug: sync_unplug: No "\n" in task name From: Yong Zhang <yong.zhang0@gmail.com> Date: Sun, 16 Oct 2011 18:56:43 +0800 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Otherwise the output will look a little odd. diff --git a/debian/patches/features/all/rt/hotplug-use-migrate-disable.patch b/debian/patches/features/all/rt/hotplug-use-migrate-disable.patch index 787b7f369b3c..5188b3f62a57 100644 --- a/debian/patches/features/all/rt/hotplug-use-migrate-disable.patch +++ b/debian/patches/features/all/rt/hotplug-use-migrate-disable.patch @@ -1,7 +1,7 @@ Subject: hotplug: Use migrate disable on unplug From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 19:35:29 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Migration needs to be disabled accross the unplug handling to make sure that the unplug thread is off the unplugged cpu. diff --git a/debian/patches/features/all/rt/hrtimer-Move-schedule_work-call-to-helper-thread.patch b/debian/patches/features/all/rt/hrtimer-Move-schedule_work-call-to-helper-thread.patch index fac558eaaea2..62ca4a7be58c 100644 --- a/debian/patches/features/all/rt/hrtimer-Move-schedule_work-call-to-helper-thread.patch +++ b/debian/patches/features/all/rt/hrtimer-Move-schedule_work-call-to-helper-thread.patch @@ -1,7 +1,7 @@ From: Yang Shi <yang.shi@windriver.com> Date: Mon, 16 Sep 2013 14:09:19 -0700 Subject: hrtimer: Move schedule_work call to helper thread -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz When run ltp leapsec_timer test, the following call trace is caught: diff --git a/debian/patches/features/all/rt/hrtimer-enfore-64byte-alignment.patch b/debian/patches/features/all/rt/hrtimer-enfore-64byte-alignment.patch index 7082b4615c15..ffc0357e7bc0 100644 --- a/debian/patches/features/all/rt/hrtimer-enfore-64byte-alignment.patch +++ b/debian/patches/features/all/rt/hrtimer-enfore-64byte-alignment.patch @@ -2,7 +2,7 @@ From e35e67cb032e78055b63eae5a3a370664fabfc01 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed, 23 Dec 2015 20:57:41 +0100 Subject: [PATCH] hrtimer: enfore 64byte alignment -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The patch "hrtimer: Fixup hrtimer callback changes for preempt-rt" adds a list_head expired to struct hrtimer_clock_base and with it we run into diff --git a/debian/patches/features/all/rt/hrtimer-fixup-hrtimer-callback-changes-for-preempt-r.patch b/debian/patches/features/all/rt/hrtimer-fixup-hrtimer-callback-changes-for-preempt-r.patch index 1e653d94d60b..d4e01ee6f037 100644 --- a/debian/patches/features/all/rt/hrtimer-fixup-hrtimer-callback-changes-for-preempt-r.patch +++ b/debian/patches/features/all/rt/hrtimer-fixup-hrtimer-callback-changes-for-preempt-r.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 3 Jul 2009 08:44:31 -0500 Subject: hrtimer: Fixup hrtimer callback changes for preempt-rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz In preempt-rt we can not call the callbacks which take sleeping locks from the timer interrupt context. diff --git a/debian/patches/features/all/rt/hrtimers-prepare-full-preemption.patch b/debian/patches/features/all/rt/hrtimers-prepare-full-preemption.patch index 40176dd6f00c..9c361630484b 100644 --- a/debian/patches/features/all/rt/hrtimers-prepare-full-preemption.patch +++ b/debian/patches/features/all/rt/hrtimers-prepare-full-preemption.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:34 -0500 Subject: hrtimers: Prepare full preemption -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Make cancellation of a running callback in softirq context safe against preemption. diff --git a/debian/patches/features/all/rt/hwlat-detector-Don-t-ignore-threshold-module-paramet.patch b/debian/patches/features/all/rt/hwlat-detector-Don-t-ignore-threshold-module-paramet.patch index 2588e46a8bdd..1ff69c0d67fd 100644 --- a/debian/patches/features/all/rt/hwlat-detector-Don-t-ignore-threshold-module-paramet.patch +++ b/debian/patches/features/all/rt/hwlat-detector-Don-t-ignore-threshold-module-paramet.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <bitbucket@online.de> Date: Fri, 30 Aug 2013 07:57:25 +0200 Subject: hwlat-detector: Don't ignore threshold module parameter -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz If the user specified a threshold at module load time, use it. diff --git a/debian/patches/features/all/rt/hwlat-detector-Update-hwlat_detector-to-add-outer-lo.patch b/debian/patches/features/all/rt/hwlat-detector-Update-hwlat_detector-to-add-outer-lo.patch index 74b8a6bd6471..2be129b8d03b 100644 --- a/debian/patches/features/all/rt/hwlat-detector-Update-hwlat_detector-to-add-outer-lo.patch +++ b/debian/patches/features/all/rt/hwlat-detector-Update-hwlat_detector-to-add-outer-lo.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Mon, 19 Aug 2013 17:33:25 -0400 Subject: hwlat-detector: Update hwlat_detector to add outer loop detection -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The hwlat_detector reads two timestamps in a row, then reports any gap between those calls. The problem is, it misses everything between diff --git a/debian/patches/features/all/rt/hwlat-detector-Use-thread-instead-of-stop-machine.patch b/debian/patches/features/all/rt/hwlat-detector-Use-thread-instead-of-stop-machine.patch index 282e9a217a45..3913d64bbcfb 100644 --- a/debian/patches/features/all/rt/hwlat-detector-Use-thread-instead-of-stop-machine.patch +++ b/debian/patches/features/all/rt/hwlat-detector-Use-thread-instead-of-stop-machine.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Mon, 19 Aug 2013 17:33:27 -0400 Subject: hwlat-detector: Use thread instead of stop machine -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz There's no reason to use stop machine to search for hardware latency. Simply disabling interrupts while running the loop will do enough to diff --git a/debian/patches/features/all/rt/hwlat-detector-Use-trace_clock_local-if-available.patch b/debian/patches/features/all/rt/hwlat-detector-Use-trace_clock_local-if-available.patch index 784f0a951ee1..dbda336cb86a 100644 --- a/debian/patches/features/all/rt/hwlat-detector-Use-trace_clock_local-if-available.patch +++ b/debian/patches/features/all/rt/hwlat-detector-Use-trace_clock_local-if-available.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Mon, 19 Aug 2013 17:33:26 -0400 Subject: hwlat-detector: Use trace_clock_local if available -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz As ktime_get() calls into the timing code which does a read_seq(), it may be affected by other CPUS that touch that lock. To remove this diff --git a/debian/patches/features/all/rt/hwlatdetect.patch b/debian/patches/features/all/rt/hwlatdetect.patch index 2345756dabae..dc4fec054104 100644 --- a/debian/patches/features/all/rt/hwlatdetect.patch +++ b/debian/patches/features/all/rt/hwlatdetect.patch @@ -1,7 +1,7 @@ Subject: hwlatdetect.patch From: Carsten Emde <C.Emde@osadl.org> Date: Tue, 19 Jul 2011 13:53:12 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Jon Masters developed this wonderful SMI detector. For details please consult Documentation/hwlat_detector.txt. It could be ported to Linux diff --git a/debian/patches/features/all/rt/i2c-omap-drop-the-lock-hard-irq-context.patch b/debian/patches/features/all/rt/i2c-omap-drop-the-lock-hard-irq-context.patch index 8324f1007010..a57322d5c5db 100644 --- a/debian/patches/features/all/rt/i2c-omap-drop-the-lock-hard-irq-context.patch +++ b/debian/patches/features/all/rt/i2c-omap-drop-the-lock-hard-irq-context.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 21 Mar 2013 11:35:49 +0100 Subject: i2c/omap: drop the lock hard irq context -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The lock is taken while reading two registers. On RT the first lock is taken in hard irq where it might sleep and in the threaded irq. diff --git a/debian/patches/features/all/rt/i915-bogus-warning-from-i915-when-running-on-PREEMPT.patch b/debian/patches/features/all/rt/i915-bogus-warning-from-i915-when-running-on-PREEMPT.patch index 8e2f6a4bf219..f1a91fb5a803 100644 --- a/debian/patches/features/all/rt/i915-bogus-warning-from-i915-when-running-on-PREEMPT.patch +++ b/debian/patches/features/all/rt/i915-bogus-warning-from-i915-when-running-on-PREEMPT.patch @@ -1,7 +1,7 @@ From: Clark Williams <williams@redhat.com> Date: Tue, 26 May 2015 10:43:43 -0500 Subject: i915: bogus warning from i915 when running on PREEMPT_RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The i915 driver has a 'WARN_ON(!in_interrupt())' in the display handler, which whines constanly on the RT kernel (since the interrupt @@ -19,7 +19,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c -@@ -11366,7 +11366,7 @@ void intel_check_page_flip(struct drm_de +@@ -11376,7 +11376,7 @@ void intel_check_page_flip(struct drm_de struct intel_crtc *intel_crtc = to_intel_crtc(crtc); struct intel_unpin_work *work; diff --git a/debian/patches/features/all/rt/i915_compile_fix.patch b/debian/patches/features/all/rt/i915_compile_fix.patch index 67cf651f1c0c..9618f2528b92 100644 --- a/debian/patches/features/all/rt/i915_compile_fix.patch +++ b/debian/patches/features/all/rt/i915_compile_fix.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Tue, 14 Jul 2015 14:26:34 +0200 Subject: gpu/i915: don't open code these things -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The opencode part is gone in 1f83fee0 ("drm/i915: clear up wedged transitions") the owner check is still there. diff --git a/debian/patches/features/all/rt/ide-use-nort-local-irq-variants.patch b/debian/patches/features/all/rt/ide-use-nort-local-irq-variants.patch index 82bbd9e0f05b..bbcc30df3cc3 100644 --- a/debian/patches/features/all/rt/ide-use-nort-local-irq-variants.patch +++ b/debian/patches/features/all/rt/ide-use-nort-local-irq-variants.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:30:16 -0500 Subject: ide: Do not disable interrupts for PREEMPT-RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use the local_irq_*_nort variants. diff --git a/debian/patches/features/all/rt/idr-use-local-lock-for-protection.patch b/debian/patches/features/all/rt/idr-use-local-lock-for-protection.patch index 3f2941ce9a45..23fb94d39473 100644 --- a/debian/patches/features/all/rt/idr-use-local-lock-for-protection.patch +++ b/debian/patches/features/all/rt/idr-use-local-lock-for-protection.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 14 Jul 2015 14:26:34 +0200 Subject: idr: Use local lock instead of preempt enable/disable -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz We need to protect the per cpu variable and prevent migration. diff --git a/debian/patches/features/all/rt/infiniband-mellanox-ib-use-nort-irq.patch b/debian/patches/features/all/rt/infiniband-mellanox-ib-use-nort-irq.patch index 7a9a05a82b7b..1299bb6234b7 100644 --- a/debian/patches/features/all/rt/infiniband-mellanox-ib-use-nort-irq.patch +++ b/debian/patches/features/all/rt/infiniband-mellanox-ib-use-nort-irq.patch @@ -1,7 +1,7 @@ From: Sven-Thorsten Dietrich <sdietrich@novell.com> Date: Fri, 3 Jul 2009 08:30:35 -0500 Subject: infiniband: Mellanox IB driver patch use _nort() primitives -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Fixes in_atomic stack-dump, when Mellanox module is loaded into the RT Kernel. diff --git a/debian/patches/features/all/rt/inpt-gameport-use-local-irq-nort.patch b/debian/patches/features/all/rt/inpt-gameport-use-local-irq-nort.patch index 0c92bd79fe09..294227d2766b 100644 --- a/debian/patches/features/all/rt/inpt-gameport-use-local-irq-nort.patch +++ b/debian/patches/features/all/rt/inpt-gameport-use-local-irq-nort.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:30:16 -0500 Subject: input: gameport: Do not disable interrupts on PREEMPT_RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use the _nort() primitives. diff --git a/debian/patches/features/all/rt/introduce_migrate_disable_cpu_light.patch b/debian/patches/features/all/rt/introduce_migrate_disable_cpu_light.patch index edc0fe4f7446..b4c5e910cd77 100644 --- a/debian/patches/features/all/rt/introduce_migrate_disable_cpu_light.patch +++ b/debian/patches/features/all/rt/introduce_migrate_disable_cpu_light.patch @@ -1,7 +1,7 @@ Subject: Intrduce migrate_disable() + cpu_light() From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 17 Jun 2011 15:42:38 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Introduce migrate_disable(). The task can't be pushed to another CPU but can be preempted. @@ -90,7 +90,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> int nr_cpus_allowed; cpumask_t cpus_allowed; -@@ -1836,9 +1842,6 @@ extern int arch_task_struct_size __read_ +@@ -1837,9 +1843,6 @@ extern int arch_task_struct_size __read_ # define arch_task_struct_size (sizeof(struct task_struct)) #endif @@ -100,7 +100,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> #define TNF_MIGRATED 0x01 #define TNF_NO_GROUP 0x02 #define TNF_SHARED 0x04 -@@ -3114,6 +3117,26 @@ static inline void set_task_cpu(struct t +@@ -3116,6 +3119,26 @@ static inline void set_task_cpu(struct t #endif /* CONFIG_SMP */ diff --git a/debian/patches/features/all/rt/ipc-msg-Implement-lockless-pipelined-wakeups.patch b/debian/patches/features/all/rt/ipc-msg-Implement-lockless-pipelined-wakeups.patch index 0f694defea1f..56665740e710 100644 --- a/debian/patches/features/all/rt/ipc-msg-Implement-lockless-pipelined-wakeups.patch +++ b/debian/patches/features/all/rt/ipc-msg-Implement-lockless-pipelined-wakeups.patch @@ -2,7 +2,7 @@ From 9a69dce752915917ecfe06a21f9c826c76f6eb07 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 30 Oct 2015 11:59:07 +0100 Subject: [PATCH] ipc/msg: Implement lockless pipelined wakeups -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz This patch moves the wakeup_process() invocation so it is not done under the perm->lock by making use of a lockless wake_q. With this change, the diff --git a/debian/patches/features/all/rt/ipc-sem-rework-semaphore-wakeups.patch b/debian/patches/features/all/rt/ipc-sem-rework-semaphore-wakeups.patch index b680526b9c12..dbd5ca70e77f 100644 --- a/debian/patches/features/all/rt/ipc-sem-rework-semaphore-wakeups.patch +++ b/debian/patches/features/all/rt/ipc-sem-rework-semaphore-wakeups.patch @@ -1,7 +1,7 @@ Subject: ipc/sem: Rework semaphore wakeups From: Peter Zijlstra <peterz@infradead.org> Date: Wed, 14 Sep 2011 11:57:04 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Current sysv sems have a weird ass wakeup scheme that involves keeping preemption disabled over a potential O(n^2) loop and busy waiting on diff --git a/debian/patches/features/all/rt/irq-allow-disabling-of-softirq-processing-in-irq-thread-context.patch b/debian/patches/features/all/rt/irq-allow-disabling-of-softirq-processing-in-irq-thread-context.patch index e10cf83292a5..0cde3db9d9cd 100644 --- a/debian/patches/features/all/rt/irq-allow-disabling-of-softirq-processing-in-irq-thread-context.patch +++ b/debian/patches/features/all/rt/irq-allow-disabling-of-softirq-processing-in-irq-thread-context.patch @@ -1,7 +1,7 @@ Subject: genirq: Allow disabling of softirq processing in irq thread context From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 31 Jan 2012 13:01:27 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The processing of softirqs in irq thread context is a performance gain for the non-rt workloads of a system, but it's counterproductive for diff --git a/debian/patches/features/all/rt/irqwork-Move-irq-safe-work-to-irq-context.patch b/debian/patches/features/all/rt/irqwork-Move-irq-safe-work-to-irq-context.patch index b11065e69ef7..21ff0e61d812 100644 --- a/debian/patches/features/all/rt/irqwork-Move-irq-safe-work-to-irq-context.patch +++ b/debian/patches/features/all/rt/irqwork-Move-irq-safe-work-to-irq-context.patch @@ -1,7 +1,7 @@ Subject: irqwork: Move irq safe work to irq context From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 15 Nov 2015 18:40:17 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On architectures where arch_irq_work_has_interrupt() returns false, we end up running the irq safe work from the softirq context. That diff --git a/debian/patches/features/all/rt/irqwork-push_most_work_into_softirq_context.patch b/debian/patches/features/all/rt/irqwork-push_most_work_into_softirq_context.patch index e00bba675a9c..eb3bf05eadfb 100644 --- a/debian/patches/features/all/rt/irqwork-push_most_work_into_softirq_context.patch +++ b/debian/patches/features/all/rt/irqwork-push_most_work_into_softirq_context.patch @@ -1,7 +1,7 @@ Subject: irqwork: push most work into softirq context From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Tue, 23 Jun 2015 15:32:51 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Initially we defered all irqwork into softirq because we didn't want the latency spikes if perf or another user was busy and delayed the RT task. diff --git a/debian/patches/features/all/rt/jump-label-rt.patch b/debian/patches/features/all/rt/jump-label-rt.patch index 5bf452f42a2c..221f57590f1b 100644 --- a/debian/patches/features/all/rt/jump-label-rt.patch +++ b/debian/patches/features/all/rt/jump-label-rt.patch @@ -1,7 +1,7 @@ Subject: jump-label: disable if stop_machine() is used From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 08 Jul 2015 17:14:48 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Some architectures are using stop_machine() while switching the opcode which leads to latency spikes. diff --git a/debian/patches/features/all/rt/kconfig-disable-a-few-options-rt.patch b/debian/patches/features/all/rt/kconfig-disable-a-few-options-rt.patch index c90f49ec6e02..7e4493e43b13 100644 --- a/debian/patches/features/all/rt/kconfig-disable-a-few-options-rt.patch +++ b/debian/patches/features/all/rt/kconfig-disable-a-few-options-rt.patch @@ -1,7 +1,7 @@ Subject: kconfig: Disable config options which are not RT compatible From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 24 Jul 2011 12:11:43 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Disable stuff which is known to have issues on RT diff --git a/debian/patches/features/all/rt/kconfig-preempt-rt-full.patch b/debian/patches/features/all/rt/kconfig-preempt-rt-full.patch index 4b46e742dfa9..25ce228ed2c8 100644 --- a/debian/patches/features/all/rt/kconfig-preempt-rt-full.patch +++ b/debian/patches/features/all/rt/kconfig-preempt-rt-full.patch @@ -1,7 +1,7 @@ Subject: kconfig: Add PREEMPT_RT_FULL From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 29 Jun 2011 14:58:57 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Introduce the final symbol for PREEMPT_RT_FULL. diff --git a/debian/patches/features/all/rt/kernel-SRCU-provide-a-static-initializer.patch b/debian/patches/features/all/rt/kernel-SRCU-provide-a-static-initializer.patch index 34ae1a59f2e7..db71b8fe0368 100644 --- a/debian/patches/features/all/rt/kernel-SRCU-provide-a-static-initializer.patch +++ b/debian/patches/features/all/rt/kernel-SRCU-provide-a-static-initializer.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Tue, 19 Mar 2013 14:44:30 +0100 Subject: [PATCH] kernel/SRCU: provide a static initializer -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz There are macros for static initializer for the three out of four possible notifier types, that are: diff --git a/debian/patches/features/all/rt/kernel-cpu-fix-cpu-down-problem-if-kthread-s-cpu-is-.patch b/debian/patches/features/all/rt/kernel-cpu-fix-cpu-down-problem-if-kthread-s-cpu-is-.patch index 3b3264d35841..264f60aaf00d 100644 --- a/debian/patches/features/all/rt/kernel-cpu-fix-cpu-down-problem-if-kthread-s-cpu-is-.patch +++ b/debian/patches/features/all/rt/kernel-cpu-fix-cpu-down-problem-if-kthread-s-cpu-is-.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 7 Jun 2013 22:37:06 +0200 Subject: kernel/cpu: fix cpu down problem if kthread's cpu is going down -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz If kthread is pinned to CPUx and CPUx is going down then we get into trouble: diff --git a/debian/patches/features/all/rt/kernel-hotplug-restore-original-cpu-mask-oncpu-down.patch b/debian/patches/features/all/rt/kernel-hotplug-restore-original-cpu-mask-oncpu-down.patch index 616e5f406fab..3a9f9d4b401b 100644 --- a/debian/patches/features/all/rt/kernel-hotplug-restore-original-cpu-mask-oncpu-down.patch +++ b/debian/patches/features/all/rt/kernel-hotplug-restore-original-cpu-mask-oncpu-down.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 14 Jun 2013 17:16:35 +0200 Subject: kernel/hotplug: restore original cpu mask oncpu/down -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz If a task which is allowed to run only on CPU X puts CPU Y down then it will be allowed on all CPUs but the on CPU Y after it comes back from diff --git a/debian/patches/features/all/rt/kgb-serial-hackaround.patch b/debian/patches/features/all/rt/kgb-serial-hackaround.patch index b903eac6d1c3..7064c4e4c675 100644 --- a/debian/patches/features/all/rt/kgb-serial-hackaround.patch +++ b/debian/patches/features/all/rt/kgb-serial-hackaround.patch @@ -1,7 +1,7 @@ From: Jason Wessel <jason.wessel@windriver.com> Date: Thu, 28 Jul 2011 12:42:23 -0500 Subject: kgdb/serial: Short term workaround -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On 07/27/2011 04:37 PM, Thomas Gleixner wrote: > - KGDB (not yet disabled) is reportedly unusable on -rt right now due diff --git a/debian/patches/features/all/rt/latency-hist.patch b/debian/patches/features/all/rt/latency-hist.patch index 88a5464c2f81..263060fddfa3 100644 --- a/debian/patches/features/all/rt/latency-hist.patch +++ b/debian/patches/features/all/rt/latency-hist.patch @@ -1,7 +1,7 @@ Subject: tracing: Add latency histograms From: Carsten Emde <C.Emde@osadl.org> Date: Tue, 19 Jul 2011 14:03:41 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz This patch provides a recording mechanism to store data of potential sources of system latencies. The recordings separately determine the @@ -229,7 +229,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> void *start_site; --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -1794,6 +1794,12 @@ struct task_struct { +@@ -1795,6 +1795,12 @@ struct task_struct { unsigned long trace; /* bitmask and counter of trace recursion */ unsigned long trace_recursion; diff --git a/debian/patches/features/all/rt/latency_hist-update-sched_wakeup-probe.patch b/debian/patches/features/all/rt/latency_hist-update-sched_wakeup-probe.patch index 9bb3e9225aa1..1a8823e9f21a 100644 --- a/debian/patches/features/all/rt/latency_hist-update-sched_wakeup-probe.patch +++ b/debian/patches/features/all/rt/latency_hist-update-sched_wakeup-probe.patch @@ -1,7 +1,7 @@ Subject: latency_hist: Update sched_wakeup probe From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Date: Sun, 25 Oct 2015 18:06:05 -0400 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz "sched: Introduce the 'trace_sched_waking' tracepoint" introduces a prototype change for the sched_wakeup probe: the "success" argument is diff --git a/debian/patches/features/all/rt/leds-trigger-disable-CPU-trigger-on-RT.patch b/debian/patches/features/all/rt/leds-trigger-disable-CPU-trigger-on-RT.patch index c7e17d303c64..b8eaba62b17e 100644 --- a/debian/patches/features/all/rt/leds-trigger-disable-CPU-trigger-on-RT.patch +++ b/debian/patches/features/all/rt/leds-trigger-disable-CPU-trigger-on-RT.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 23 Jan 2014 14:45:59 +0100 Subject: leds: trigger: disable CPU trigger on -RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz as it triggers: |CPU: 0 PID: 0 Comm: swapper Not tainted 3.12.8-rt10 #141 diff --git a/debian/patches/features/all/rt/lglocks-rt.patch b/debian/patches/features/all/rt/lglocks-rt.patch index b225459ef392..76712342a483 100644 --- a/debian/patches/features/all/rt/lglocks-rt.patch +++ b/debian/patches/features/all/rt/lglocks-rt.patch @@ -1,7 +1,7 @@ Subject: lglocks: Provide a RT safe variant From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 15 Jun 2011 11:02:21 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz lglocks by itself will spin in order to get the lock. This will end up badly if a task with the highest priority keeps spinning while a task diff --git a/debian/patches/features/all/rt/list_bl.h-make-list-head-locking-RT-safe.patch b/debian/patches/features/all/rt/list_bl.h-make-list-head-locking-RT-safe.patch index d3314addc81b..1ac90957e80d 100644 --- a/debian/patches/features/all/rt/list_bl.h-make-list-head-locking-RT-safe.patch +++ b/debian/patches/features/all/rt/list_bl.h-make-list-head-locking-RT-safe.patch @@ -1,7 +1,7 @@ From: Paul Gortmaker <paul.gortmaker@windriver.com> Date: Fri, 21 Jun 2013 15:07:25 -0400 Subject: list_bl: Make list head locking RT safe -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz As per changes in include/linux/jbd_common.h for avoiding the bit_spin_locks on RT ("fs: jbd/jbd2: Make state lock and journal diff --git a/debian/patches/features/all/rt/local-irq-rt-depending-variants.patch b/debian/patches/features/all/rt/local-irq-rt-depending-variants.patch index 746d3f0d7a7a..9e2fd38ba723 100644 --- a/debian/patches/features/all/rt/local-irq-rt-depending-variants.patch +++ b/debian/patches/features/all/rt/local-irq-rt-depending-variants.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 21 Jul 2009 22:34:14 +0200 Subject: rt: local_irq_* variants depending on RT/!RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Add local_irq_*_(no)rt variant which are mainly used to break interrupt disabled sections on PREEMPT_RT or to explicitely disable diff --git a/debian/patches/features/all/rt/localversion.patch b/debian/patches/features/all/rt/localversion.patch index 11f2fc54f224..e2813c278f68 100644 --- a/debian/patches/features/all/rt/localversion.patch +++ b/debian/patches/features/all/rt/localversion.patch @@ -1,7 +1,7 @@ -Subject: v4.4-rc6-rt1 +Subject: v4.4-rt2 From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 08 Jul 2011 20:25:16 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- @@ -11,4 +11,4 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- /dev/null +++ b/localversion-rt @@ -0,0 +1 @@ -+-rt1 ++-rt2 diff --git a/debian/patches/features/all/rt/lockdep-no-softirq-accounting-on-rt.patch b/debian/patches/features/all/rt/lockdep-no-softirq-accounting-on-rt.patch index e97d52fdb8c8..bc95f1c9ef82 100644 --- a/debian/patches/features/all/rt/lockdep-no-softirq-accounting-on-rt.patch +++ b/debian/patches/features/all/rt/lockdep-no-softirq-accounting-on-rt.patch @@ -1,7 +1,7 @@ Subject: lockdep: Make it RT aware From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 18:51:23 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz teach lockdep that we don't really do softirqs on -RT. diff --git a/debian/patches/features/all/rt/lockdep-selftest-fix-warnings-due-to-missing-PREEMPT.patch b/debian/patches/features/all/rt/lockdep-selftest-fix-warnings-due-to-missing-PREEMPT.patch index 29a5f79861d6..da777f134984 100644 --- a/debian/patches/features/all/rt/lockdep-selftest-fix-warnings-due-to-missing-PREEMPT.patch +++ b/debian/patches/features/all/rt/lockdep-selftest-fix-warnings-due-to-missing-PREEMPT.patch @@ -1,7 +1,7 @@ From: Josh Cartwright <josh.cartwright@ni.com> Date: Wed, 28 Jan 2015 13:08:45 -0600 Subject: lockdep: selftest: fix warnings due to missing PREEMPT_RT conditionals -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz "lockdep: Selftest: Only do hardirq context test for raw spinlock" disabled the execution of certain tests with PREEMPT_RT_FULL, but did diff --git a/debian/patches/features/all/rt/lockdep-selftest-only-do-hardirq-context-test-for-raw-spinlock.patch b/debian/patches/features/all/rt/lockdep-selftest-only-do-hardirq-context-test-for-raw-spinlock.patch index c796a8b4f10c..134d2a36e27c 100644 --- a/debian/patches/features/all/rt/lockdep-selftest-only-do-hardirq-context-test-for-raw-spinlock.patch +++ b/debian/patches/features/all/rt/lockdep-selftest-only-do-hardirq-context-test-for-raw-spinlock.patch @@ -1,7 +1,7 @@ Subject: lockdep: selftest: Only do hardirq context test for raw spinlock From: Yong Zhang <yong.zhang0@gmail.com> Date: Mon, 16 Apr 2012 15:01:56 +0800 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz From: Yong Zhang <yong.zhang@windriver.com> diff --git a/debian/patches/features/all/rt/locking-locktorture-Do-NOT-include-rwlock.h-directly.patch b/debian/patches/features/all/rt/locking-locktorture-Do-NOT-include-rwlock.h-directly.patch index f3b2c4b139f9..0ff8f53dd20e 100644 --- a/debian/patches/features/all/rt/locking-locktorture-Do-NOT-include-rwlock.h-directly.patch +++ b/debian/patches/features/all/rt/locking-locktorture-Do-NOT-include-rwlock.h-directly.patch @@ -1,7 +1,7 @@ From: "Wolfgang M. Reimer" <linuxball@gmail.com> Date: Tue, 21 Jul 2015 16:20:07 +0200 Subject: locking: locktorture: Do NOT include rwlock.h directly -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Including rwlock.h directly will cause kernel builds to fail if CONFIG_PREEMPT_RT_FULL is defined. The correct header file diff --git a/debian/patches/features/all/rt/md-disable-bcache.patch b/debian/patches/features/all/rt/md-disable-bcache.patch index 89e7d35a7b97..a3a260a629d5 100644 --- a/debian/patches/features/all/rt/md-disable-bcache.patch +++ b/debian/patches/features/all/rt/md-disable-bcache.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 29 Aug 2013 11:48:57 +0200 Subject: md: disable bcache -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz It uses anon semaphores |drivers/md/bcache/request.c: In function ‘cached_dev_write_complete’: diff --git a/debian/patches/features/all/rt/md-raid5-percpu-handling-rt-aware.patch b/debian/patches/features/all/rt/md-raid5-percpu-handling-rt-aware.patch index b2d08fbd70d8..d2b3924eb139 100644 --- a/debian/patches/features/all/rt/md-raid5-percpu-handling-rt-aware.patch +++ b/debian/patches/features/all/rt/md-raid5-percpu-handling-rt-aware.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 6 Apr 2010 16:51:31 +0200 Subject: md: raid5: Make raid5_percpu handling RT aware -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz __raid_run_ops() disables preemption with get_cpu() around the access to the raid5_percpu variables. That causes scheduling while atomic diff --git a/debian/patches/features/all/rt/mips-disable-highmem-on-rt.patch b/debian/patches/features/all/rt/mips-disable-highmem-on-rt.patch index 92600f2d70c5..e87ebda79676 100644 --- a/debian/patches/features/all/rt/mips-disable-highmem-on-rt.patch +++ b/debian/patches/features/all/rt/mips-disable-highmem-on-rt.patch @@ -1,7 +1,7 @@ Subject: mips: Disable highmem on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 18 Jul 2011 17:10:12 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The current highmem handling on -RT is not compatible and needs fixups. diff --git a/debian/patches/features/all/rt/mm--rt--Fix-generic-kmap_atomic-for-RT b/debian/patches/features/all/rt/mm--rt--Fix-generic-kmap_atomic-for-RT index a7d5988d5ae6..50de786a6591 100644 --- a/debian/patches/features/all/rt/mm--rt--Fix-generic-kmap_atomic-for-RT +++ b/debian/patches/features/all/rt/mm--rt--Fix-generic-kmap_atomic-for-RT @@ -1,7 +1,7 @@ Subject: mm: rt: Fix generic kmap_atomic for RT From: Thomas Gleixner <tglx@linutronix.de> Date: Sat, 19 Sep 2015 10:15:00 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The update to 4.1 brought in the mainline variant of the pagefault disable distangling from preempt count. That introduced a diff --git a/debian/patches/features/all/rt/mm-bounce-local-irq-save-nort.patch b/debian/patches/features/all/rt/mm-bounce-local-irq-save-nort.patch index 077ed028a268..a7b22c436aa2 100644 --- a/debian/patches/features/all/rt/mm-bounce-local-irq-save-nort.patch +++ b/debian/patches/features/all/rt/mm-bounce-local-irq-save-nort.patch @@ -1,7 +1,7 @@ Subject: mm: bounce: Use local_irq_save_nort From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 09 Jan 2013 10:33:09 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz kmap_atomic() is preemptible on RT. diff --git a/debian/patches/features/all/rt/mm-convert-swap-to-percpu-locked.patch b/debian/patches/features/all/rt/mm-convert-swap-to-percpu-locked.patch index e5fc6222fb5d..a1e4cb1f8be3 100644 --- a/debian/patches/features/all/rt/mm-convert-swap-to-percpu-locked.patch +++ b/debian/patches/features/all/rt/mm-convert-swap-to-percpu-locked.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:51 -0500 Subject: mm/swap: Convert to percpu locked -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Replace global locks (get_cpu + local_irq_save) with "local_locks()". Currently there is one of for "rotate" and one for "swap". diff --git a/debian/patches/features/all/rt/mm-disable-sloub-rt.patch b/debian/patches/features/all/rt/mm-disable-sloub-rt.patch index c0acf46cbe7f..c02e71abdee4 100644 --- a/debian/patches/features/all/rt/mm-disable-sloub-rt.patch +++ b/debian/patches/features/all/rt/mm-disable-sloub-rt.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:44:03 -0500 Subject: mm: Allow only slub on RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Disable SLAB and SLOB on -RT. Only SLUB is adopted to -RT needs. diff --git a/debian/patches/features/all/rt/mm-enable-slub.patch b/debian/patches/features/all/rt/mm-enable-slub.patch index 3249884fcac2..c1f063ee24da 100644 --- a/debian/patches/features/all/rt/mm-enable-slub.patch +++ b/debian/patches/features/all/rt/mm-enable-slub.patch @@ -1,7 +1,7 @@ Subject: mm: Enable SLUB for RT From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 25 Oct 2012 10:32:35 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Make SLUB RT aware by converting locks to raw and using free lists to move the freeing out of the lock held region. diff --git a/debian/patches/features/all/rt/mm-make-vmstat-rt-aware.patch b/debian/patches/features/all/rt/mm-make-vmstat-rt-aware.patch index 2bb9951e8a87..908b90ef905c 100644 --- a/debian/patches/features/all/rt/mm-make-vmstat-rt-aware.patch +++ b/debian/patches/features/all/rt/mm-make-vmstat-rt-aware.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:30:13 -0500 Subject: mm/vmstat: Protect per cpu variables with preempt disable on RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Disable preemption on -RT for the vmstat code. On vanila the code runs in IRQ-off regions while on -RT it is not. "preempt_disable" ensures that the diff --git a/debian/patches/features/all/rt/mm-memcontrol-Don-t-call-schedule_work_on-in-preempt.patch b/debian/patches/features/all/rt/mm-memcontrol-Don-t-call-schedule_work_on-in-preempt.patch index 940e11b9fd21..a7662c0deb96 100644 --- a/debian/patches/features/all/rt/mm-memcontrol-Don-t-call-schedule_work_on-in-preempt.patch +++ b/debian/patches/features/all/rt/mm-memcontrol-Don-t-call-schedule_work_on-in-preempt.patch @@ -1,7 +1,7 @@ From: Yang Shi <yang.shi@windriver.com> Subject: mm/memcontrol: Don't call schedule_work_on in preemption disabled context Date: Wed, 30 Oct 2013 11:48:33 -0700 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The following trace is triggered when running ltp oom test cases: @@ -49,7 +49,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> --- a/mm/memcontrol.c +++ b/mm/memcontrol.c -@@ -1933,7 +1933,7 @@ static void drain_all_stock(struct mem_c +@@ -1957,7 +1957,7 @@ static void drain_all_stock(struct mem_c return; /* Notify other cpus that system-wide "drain" is running */ get_online_cpus(); @@ -58,7 +58,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> for_each_online_cpu(cpu) { struct memcg_stock_pcp *stock = &per_cpu(memcg_stock, cpu); struct mem_cgroup *memcg; -@@ -1950,7 +1950,7 @@ static void drain_all_stock(struct mem_c +@@ -1974,7 +1974,7 @@ static void drain_all_stock(struct mem_c schedule_work_on(cpu, &stock->work); } } diff --git a/debian/patches/features/all/rt/mm-memcontrol-do_not_disable_irq.patch b/debian/patches/features/all/rt/mm-memcontrol-do_not_disable_irq.patch index 03c04115c464..848bc85d16bb 100644 --- a/debian/patches/features/all/rt/mm-memcontrol-do_not_disable_irq.patch +++ b/debian/patches/features/all/rt/mm-memcontrol-do_not_disable_irq.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Subject: mm/memcontrol: Replace local_irq_disable with local locks Date: Wed, 28 Jan 2015 17:14:16 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz There are a few local_irq_disable() which then take sleeping locks. This patch converts them local locks. @@ -60,7 +60,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> static const char * const mem_cgroup_stat_names[] = { "cache", "rss", -@@ -4584,12 +4587,12 @@ static int mem_cgroup_move_account(struc +@@ -4615,12 +4618,12 @@ static int mem_cgroup_move_account(struc ret = 0; @@ -75,7 +75,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> out_unlock: unlock_page(page); out: -@@ -5341,10 +5344,10 @@ void mem_cgroup_commit_charge(struct pag +@@ -5373,10 +5376,10 @@ void mem_cgroup_commit_charge(struct pag VM_BUG_ON_PAGE(!PageTransHuge(page), page); } @@ -88,7 +88,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> if (do_swap_account && PageSwapCache(page)) { swp_entry_t entry = { .val = page_private(page) }; -@@ -5400,14 +5403,14 @@ static void uncharge_batch(struct mem_cg +@@ -5432,14 +5435,14 @@ static void uncharge_batch(struct mem_cg memcg_oom_recover(memcg); } @@ -105,7 +105,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> if (!mem_cgroup_is_root(memcg)) css_put_many(&memcg->css, nr_pages); -@@ -5599,6 +5602,7 @@ void mem_cgroup_swapout(struct page *pag +@@ -5631,6 +5634,7 @@ void mem_cgroup_swapout(struct page *pag { struct mem_cgroup *memcg; unsigned short oldid; @@ -113,7 +113,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> VM_BUG_ON_PAGE(PageLRU(page), page); VM_BUG_ON_PAGE(page_count(page), page); -@@ -5627,9 +5631,13 @@ void mem_cgroup_swapout(struct page *pag +@@ -5659,9 +5663,13 @@ void mem_cgroup_swapout(struct page *pag * important here to have the interrupts disabled because it is the * only synchronisation we have for udpating the per-CPU variables. */ diff --git a/debian/patches/features/all/rt/mm-page-alloc-use-local-lock-on-target-cpu.patch b/debian/patches/features/all/rt/mm-page-alloc-use-local-lock-on-target-cpu.patch index 4a52b426cda7..aa1b56b853ad 100644 --- a/debian/patches/features/all/rt/mm-page-alloc-use-local-lock-on-target-cpu.patch +++ b/debian/patches/features/all/rt/mm-page-alloc-use-local-lock-on-target-cpu.patch @@ -1,7 +1,7 @@ Subject: mm: page_alloc: Use local_lock_on() instead of plain spinlock From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 27 Sep 2012 11:11:46 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The plain spinlock while sufficient does not update the local_lock internals. Use a proper local_lock function instead to ease debugging. diff --git a/debian/patches/features/all/rt/mm-page_alloc-reduce-lock-sections-further.patch b/debian/patches/features/all/rt/mm-page_alloc-reduce-lock-sections-further.patch index cda3a103e3fe..380c888d8f07 100644 --- a/debian/patches/features/all/rt/mm-page_alloc-reduce-lock-sections-further.patch +++ b/debian/patches/features/all/rt/mm-page_alloc-reduce-lock-sections-further.patch @@ -1,7 +1,7 @@ From: Peter Zijlstra <peterz@infradead.org> Date: Fri Jul 3 08:44:37 2009 -0500 Subject: mm: page_alloc: Reduce lock sections further -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Split out the pages which are to be freed into a separate list and call free_pages_bulk() outside of the percpu page allocator locks. diff --git a/debian/patches/features/all/rt/mm-page_alloc-rt-friendly-per-cpu-pages.patch b/debian/patches/features/all/rt/mm-page_alloc-rt-friendly-per-cpu-pages.patch index 02e2bb59e4ea..e8c157414880 100644 --- a/debian/patches/features/all/rt/mm-page_alloc-rt-friendly-per-cpu-pages.patch +++ b/debian/patches/features/all/rt/mm-page_alloc-rt-friendly-per-cpu-pages.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:37 -0500 Subject: mm: page_alloc: rt-friendly per-cpu pages -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz rt-friendly per-cpu pages: convert the irqs-off per-cpu locking method into a preemptible, explicit-per-cpu-locks method. diff --git a/debian/patches/features/all/rt/mm-protect-activate-switch-mm.patch b/debian/patches/features/all/rt/mm-protect-activate-switch-mm.patch index d8ad8f2b2ff3..2e74137ec30b 100644 --- a/debian/patches/features/all/rt/mm-protect-activate-switch-mm.patch +++ b/debian/patches/features/all/rt/mm-protect-activate-switch-mm.patch @@ -1,7 +1,7 @@ From: Yong Zhang <yong.zhang0@gmail.com> Date: Tue, 15 May 2012 13:53:56 +0800 Subject: mm: Protect activate_mm() by preempt_[disable&enable]_rt() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz User preempt_*_rt instead of local_irq_*_rt or otherwise there will be warning on ARM like below: diff --git a/debian/patches/features/all/rt/mm-rmap-retry-lock-check-in-anon_vma_free.patch_vma_free.patch b/debian/patches/features/all/rt/mm-rmap-retry-lock-check-in-anon_vma_free.patch_vma_free.patch index d1adc2850c2d..c8d7b7dba6f1 100644 --- a/debian/patches/features/all/rt/mm-rmap-retry-lock-check-in-anon_vma_free.patch_vma_free.patch +++ b/debian/patches/features/all/rt/mm-rmap-retry-lock-check-in-anon_vma_free.patch_vma_free.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Tue, 1 Dec 2015 17:57:02 +0100 Subject: mm/rmap: retry lock check in anon_vma_free() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz anon_vma_free() checks if the rwsem is locked and if so performs a rw lock + unlock operation. It seems the purpose is to flush the current diff --git a/debian/patches/features/all/rt/mm-rt-kmap-atomic-scheduling.patch b/debian/patches/features/all/rt/mm-rt-kmap-atomic-scheduling.patch index 359465ca2cab..90121b42b1f3 100644 --- a/debian/patches/features/all/rt/mm-rt-kmap-atomic-scheduling.patch +++ b/debian/patches/features/all/rt/mm-rt-kmap-atomic-scheduling.patch @@ -1,7 +1,7 @@ Subject: mm, rt: kmap_atomic scheduling From: Peter Zijlstra <peterz@infradead.org> Date: Thu, 28 Jul 2011 10:43:51 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz In fact, with migrate_disable() existing one could play games with kmap_atomic. You could save/restore the kmap_atomic slots on context @@ -230,7 +230,7 @@ Link: http://lkml.kernel.org/r/1311842631.5890.208.camel@twins #include <asm/page.h> #include <asm/ptrace.h> -@@ -1847,6 +1848,12 @@ struct task_struct { +@@ -1848,6 +1849,12 @@ struct task_struct { int softirq_nestcnt; unsigned int softirqs_raised; #endif diff --git a/debian/patches/features/all/rt/mm-scatterlist-dont-disable-irqs-on-RT.patch b/debian/patches/features/all/rt/mm-scatterlist-dont-disable-irqs-on-RT.patch index 51a5b2efd237..06bd7ddba510 100644 --- a/debian/patches/features/all/rt/mm-scatterlist-dont-disable-irqs-on-RT.patch +++ b/debian/patches/features/all/rt/mm-scatterlist-dont-disable-irqs-on-RT.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 3 Jul 2009 08:44:34 -0500 Subject: mm/scatterlist: Do not disable irqs on RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The local_irq_save() is not only used to get things done "fast" but also to ensure that in case of SG_MITER_ATOMIC we are in "atomic" diff --git a/debian/patches/features/all/rt/mm-vmalloc-use-get-cpu-light.patch b/debian/patches/features/all/rt/mm-vmalloc-use-get-cpu-light.patch index 72ab4159e755..766ddfca91cf 100644 --- a/debian/patches/features/all/rt/mm-vmalloc-use-get-cpu-light.patch +++ b/debian/patches/features/all/rt/mm-vmalloc-use-get-cpu-light.patch @@ -1,7 +1,7 @@ Subject: mm/vmalloc: Another preempt disable region which sucks From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 12 Jul 2011 11:39:36 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Avoid the preempt disable version of get_cpu_var(). The inner-lock should provide enough serialisation. diff --git a/debian/patches/features/all/rt/mm-workingset-do-not-protect-workingset_shadow_nodes.patch b/debian/patches/features/all/rt/mm-workingset-do-not-protect-workingset_shadow_nodes.patch index 86e690f6f720..11323abe71e6 100644 --- a/debian/patches/features/all/rt/mm-workingset-do-not-protect-workingset_shadow_nodes.patch +++ b/debian/patches/features/all/rt/mm-workingset-do-not-protect-workingset_shadow_nodes.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 29 Jan 2015 17:19:44 +0100 Subject: mm/workingset: Do not protect workingset_shadow_nodes with irq off -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz workingset_shadow_nodes is protected by local_irq_disable(). Some users use spin_lock_irq(). diff --git a/debian/patches/features/all/rt/mmci-remove-bogus-irq-save.patch b/debian/patches/features/all/rt/mmci-remove-bogus-irq-save.patch index 04fa1aab27f4..43985ec1b596 100644 --- a/debian/patches/features/all/rt/mmci-remove-bogus-irq-save.patch +++ b/debian/patches/features/all/rt/mmci-remove-bogus-irq-save.patch @@ -1,7 +1,7 @@ Subject: mmci: Remove bogus local_irq_save() From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 09 Jan 2013 12:11:12 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On !RT interrupt runs with interrupts disabled. On RT it's in a thread, so no need to disable interrupts at all. diff --git a/debian/patches/features/all/rt/move_sched_delayed_work_to_helper.patch b/debian/patches/features/all/rt/move_sched_delayed_work_to_helper.patch index bb3d7f1e6d76..80a3839753d1 100644 --- a/debian/patches/features/all/rt/move_sched_delayed_work_to_helper.patch +++ b/debian/patches/features/all/rt/move_sched_delayed_work_to_helper.patch @@ -1,7 +1,7 @@ Date: Wed, 26 Jun 2013 15:28:11 -0400 From: Steven Rostedt <rostedt@goodmis.org> Subject: rt,ntp: Move call to schedule_delayed_work() to helper thread -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The ntp code for notify_cmos_timer() is called from a hard interrupt context. schedule_delayed_work() under PREEMPT_RT_FULL calls spinlocks diff --git a/debian/patches/features/all/rt/mutex-no-spin-on-rt.patch b/debian/patches/features/all/rt/mutex-no-spin-on-rt.patch index 06d350bcc5f0..860c6c8ce219 100644 --- a/debian/patches/features/all/rt/mutex-no-spin-on-rt.patch +++ b/debian/patches/features/all/rt/mutex-no-spin-on-rt.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 21:51:45 +0200 Subject: locking: Disable spin on owner for RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Drop spin on owner for mutex / rwsem. We are most likely not using it but… diff --git a/debian/patches/features/all/rt/net-another-local-irq-disable-alloc-atomic-headache.patch b/debian/patches/features/all/rt/net-another-local-irq-disable-alloc-atomic-headache.patch index f2995a333ea4..dd6ffbb5a544 100644 --- a/debian/patches/features/all/rt/net-another-local-irq-disable-alloc-atomic-headache.patch +++ b/debian/patches/features/all/rt/net-another-local-irq-disable-alloc-atomic-headache.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 26 Sep 2012 16:21:08 +0200 Subject: net: Another local_irq_disable/kmalloc headache -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Replace it by a local lock. Though that's pretty inefficient :( diff --git a/debian/patches/features/all/rt/net-core-cpuhotplug-drain-input_pkt_queue-lockless.patch b/debian/patches/features/all/rt/net-core-cpuhotplug-drain-input_pkt_queue-lockless.patch index df03e0043fe2..90079f9bc165 100644 --- a/debian/patches/features/all/rt/net-core-cpuhotplug-drain-input_pkt_queue-lockless.patch +++ b/debian/patches/features/all/rt/net-core-cpuhotplug-drain-input_pkt_queue-lockless.patch @@ -1,7 +1,7 @@ Subject: net/core/cpuhotplug: Drain input_pkt_queue lockless From: Grygorii Strashko <grygorii.strashko@ti.com> Date: Fri, 9 Oct 2015 09:25:49 -0500 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz I can constantly see below error report with 4.1 RT-kernel on TI ARM dra7-evm if I'm trying to unplug cpu1: diff --git a/debian/patches/features/all/rt/net-fix-iptable-xt-write-recseq-begin-rt-fallout.patch b/debian/patches/features/all/rt/net-fix-iptable-xt-write-recseq-begin-rt-fallout.patch index 086c9863dd8b..dc11ad4b85f8 100644 --- a/debian/patches/features/all/rt/net-fix-iptable-xt-write-recseq-begin-rt-fallout.patch +++ b/debian/patches/features/all/rt/net-fix-iptable-xt-write-recseq-begin-rt-fallout.patch @@ -1,7 +1,7 @@ Subject: net: netfilter: Serialize xt_write_recseq sections on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 28 Oct 2012 11:18:08 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The netfilter code relies only on the implicit semantics of local_bh_disable() for serializing wt_write_recseq sections. RT breaks diff --git a/debian/patches/features/all/rt/net-make-devnet_rename_seq-a-mutex.patch b/debian/patches/features/all/rt/net-make-devnet_rename_seq-a-mutex.patch index 6609c6d384d9..e818f4e4287b 100644 --- a/debian/patches/features/all/rt/net-make-devnet_rename_seq-a-mutex.patch +++ b/debian/patches/features/all/rt/net-make-devnet_rename_seq-a-mutex.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed, 20 Mar 2013 18:06:20 +0100 Subject: net: Add a mutex around devnet_rename_seq -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On RT write_seqcount_begin() disables preemption and device_rename() allocates memory with GFP_KERNEL and grabs later the sysfs_mutex diff --git a/debian/patches/features/all/rt/net-prevent-abba-deadlock.patch b/debian/patches/features/all/rt/net-prevent-abba-deadlock.patch index 73118734c481..8d0731cf19a5 100644 --- a/debian/patches/features/all/rt/net-prevent-abba-deadlock.patch +++ b/debian/patches/features/all/rt/net-prevent-abba-deadlock.patch @@ -1,7 +1,7 @@ Subject: net-flip-lock-dep-thingy.patch From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 28 Jun 2011 10:59:58 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz ======================================================= [ INFO: possible circular locking dependency detected ] diff --git a/debian/patches/features/all/rt/net-sched-dev_deactivate_many-use-msleep-1-instead-o.patch b/debian/patches/features/all/rt/net-sched-dev_deactivate_many-use-msleep-1-instead-o.patch index eaf1ab9b05a6..32964f7f2e18 100644 --- a/debian/patches/features/all/rt/net-sched-dev_deactivate_many-use-msleep-1-instead-o.patch +++ b/debian/patches/features/all/rt/net-sched-dev_deactivate_many-use-msleep-1-instead-o.patch @@ -1,7 +1,7 @@ From: Marc Kleine-Budde <mkl@pengutronix.de> Date: Wed, 5 Mar 2014 00:49:47 +0100 Subject: net: sched: Use msleep() instead of yield() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On PREEMPT_RT enabled systems the interrupt handler run as threads at prio 50 (by default). If a high priority userspace process tries to shut down a busy @@ -47,7 +47,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c -@@ -888,7 +888,7 @@ void dev_deactivate_many(struct list_hea +@@ -890,7 +890,7 @@ void dev_deactivate_many(struct list_hea /* Wait for outstanding qdisc_run calls. */ list_for_each_entry(dev, head, close_list) while (some_qdisc_is_busy(dev)) diff --git a/debian/patches/features/all/rt/net-tx-action-avoid-livelock-on-rt.patch b/debian/patches/features/all/rt/net-tx-action-avoid-livelock-on-rt.patch index 70b8ca1245e7..5736e1d34d8d 100644 --- a/debian/patches/features/all/rt/net-tx-action-avoid-livelock-on-rt.patch +++ b/debian/patches/features/all/rt/net-tx-action-avoid-livelock-on-rt.patch @@ -1,7 +1,7 @@ Subject: net: Avoid livelock in net_tx_action() on RT From: Steven Rostedt <srostedt@redhat.com> Date: Thu, 06 Oct 2011 10:48:39 -0400 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz qdisc_lock is taken w/o disabling interrupts or bottom halfs. So code holding a qdisc_lock() can be interrupted and softirqs can run on the diff --git a/debian/patches/features/all/rt/net-use-cpu-chill.patch b/debian/patches/features/all/rt/net-use-cpu-chill.patch index e8f2db5c15d1..e726a0eaa0cb 100644 --- a/debian/patches/features/all/rt/net-use-cpu-chill.patch +++ b/debian/patches/features/all/rt/net-use-cpu-chill.patch @@ -1,7 +1,7 @@ Subject: net: Use cpu_chill() instead of cpu_relax() From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 07 Mar 2012 21:10:04 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Retry loops on RT might loop forever when the modifying side was preempted. Use cpu_chill() instead of cpu_relax() to let the system diff --git a/debian/patches/features/all/rt/net-wireless-warn-nort.patch b/debian/patches/features/all/rt/net-wireless-warn-nort.patch index 6716060895f7..a6ec0de33c25 100644 --- a/debian/patches/features/all/rt/net-wireless-warn-nort.patch +++ b/debian/patches/features/all/rt/net-wireless-warn-nort.patch @@ -1,7 +1,7 @@ Subject: net/wireless: Use WARN_ON_NORT() From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 21 Jul 2011 21:05:33 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The softirq counter is meaningless on RT, so the check triggers a false positive. diff --git a/debian/patches/features/all/rt/net__Make_synchronize-rcu_expedited_conditional-on-non-rt b/debian/patches/features/all/rt/net__Make_synchronize-rcu_expedited_conditional-on-non-rt index 0c71f82b1df5..37ccb0927847 100644 --- a/debian/patches/features/all/rt/net__Make_synchronize-rcu_expedited_conditional-on-non-rt +++ b/debian/patches/features/all/rt/net__Make_synchronize-rcu_expedited_conditional-on-non-rt @@ -1,7 +1,7 @@ Date: Tue, 27 Oct 2015 07:31:53 -0500 From: Josh Cartwright <joshc@ni.com> Subject: net: Make synchronize_rcu_expedited() conditional on !RT_FULL -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz While the use of synchronize_rcu_expedited() might make synchronize_net() "faster", it does so at significant cost on RT diff --git a/debian/patches/features/all/rt/oleg-signal-rt-fix.patch b/debian/patches/features/all/rt/oleg-signal-rt-fix.patch index b1b97765f78c..5d7001b1521c 100644 --- a/debian/patches/features/all/rt/oleg-signal-rt-fix.patch +++ b/debian/patches/features/all/rt/oleg-signal-rt-fix.patch @@ -1,7 +1,7 @@ From: Oleg Nesterov <oleg@redhat.com> Date: Tue, 14 Jul 2015 14:26:34 +0200 Subject: signal/x86: Delay calling signals in atomic -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On x86_64 we must disable preemption before we enable interrupts for stack faults, int3 and debugging, because the current task is using @@ -77,7 +77,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> #endif --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -1577,6 +1577,10 @@ struct task_struct { +@@ -1578,6 +1578,10 @@ struct task_struct { sigset_t blocked, real_blocked; sigset_t saved_sigmask; /* restored if set_restore_sigmask() was used */ struct sigpending pending; diff --git a/debian/patches/features/all/rt/panic-disable-random-on-rt.patch b/debian/patches/features/all/rt/panic-disable-random-on-rt.patch index 1d65daf6dc55..c153092d8fe2 100644 --- a/debian/patches/features/all/rt/panic-disable-random-on-rt.patch +++ b/debian/patches/features/all/rt/panic-disable-random-on-rt.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 14 Jul 2015 14:26:34 +0200 Subject: panic: skip get_random_bytes for RT_FULL in init_oops_id -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Disable on -RT. If this is invoked from irq-context we will have problems to acquire the sleeping lock. diff --git a/debian/patches/features/all/rt/patch-to-introduce-rcu-bh-qs-where-safe-from-softirq.patch b/debian/patches/features/all/rt/patch-to-introduce-rcu-bh-qs-where-safe-from-softirq.patch index 3ba3ec6b4c95..4afb1536e4c7 100644 --- a/debian/patches/features/all/rt/patch-to-introduce-rcu-bh-qs-where-safe-from-softirq.patch +++ b/debian/patches/features/all/rt/patch-to-introduce-rcu-bh-qs-where-safe-from-softirq.patch @@ -1,7 +1,7 @@ Subject: rcu: Make ksoftirqd do RCU quiescent states From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Date: Wed, 5 Oct 2011 11:45:18 -0700 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Implementing RCU-bh in terms of RCU-preempt makes the system vulnerable to network-based denial-of-service attacks. This patch therefore diff --git a/debian/patches/features/all/rt/pci-access-use-__wake_up_all_locked.patch b/debian/patches/features/all/rt/pci-access-use-__wake_up_all_locked.patch index e90cae1b9b2a..d79ec94f586c 100644 --- a/debian/patches/features/all/rt/pci-access-use-__wake_up_all_locked.patch +++ b/debian/patches/features/all/rt/pci-access-use-__wake_up_all_locked.patch @@ -1,7 +1,7 @@ Subject: pci: Use __wake_up_all_locked in pci_unblock_user_cfg_access() From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 01 Dec 2011 00:07:16 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The waitqueue is protected by the pci_lock, so we can just avoid to lock the waitqueue lock itself. That prevents the diff --git a/debian/patches/features/all/rt/percpu_ida-use-locklocks.patch b/debian/patches/features/all/rt/percpu_ida-use-locklocks.patch index 361f23441446..666e88f4e372 100644 --- a/debian/patches/features/all/rt/percpu_ida-use-locklocks.patch +++ b/debian/patches/features/all/rt/percpu_ida-use-locklocks.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed, 9 Apr 2014 11:58:17 +0200 Subject: percpu_ida: Use local locks -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz the local_irq_save() + spin_lock() does not work that well on -RT diff --git a/debian/patches/features/all/rt/perf-make-swevent-hrtimer-irqsafe.patch b/debian/patches/features/all/rt/perf-make-swevent-hrtimer-irqsafe.patch index db357956dad7..1e2c2a53731a 100644 --- a/debian/patches/features/all/rt/perf-make-swevent-hrtimer-irqsafe.patch +++ b/debian/patches/features/all/rt/perf-make-swevent-hrtimer-irqsafe.patch @@ -1,7 +1,7 @@ From: Yong Zhang <yong.zhang@windriver.com> Date: Wed, 11 Jul 2012 22:05:21 +0000 Subject: perf: Make swevent hrtimer run in irq instead of softirq -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Otherwise we get a deadlock like below: @@ -59,7 +59,7 @@ Signed-off-by: Steven Rostedt <rostedt@goodmis.org> --- a/kernel/events/core.c +++ b/kernel/events/core.c -@@ -7243,6 +7243,7 @@ static void perf_swevent_init_hrtimer(st +@@ -7228,6 +7228,7 @@ static void perf_swevent_init_hrtimer(st hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); hwc->hrtimer.function = perf_swevent_hrtimer; diff --git a/debian/patches/features/all/rt/peter_zijlstra-frob-rcu.patch b/debian/patches/features/all/rt/peter_zijlstra-frob-rcu.patch index 2fb1f9b70b14..af0d1f9acd45 100644 --- a/debian/patches/features/all/rt/peter_zijlstra-frob-rcu.patch +++ b/debian/patches/features/all/rt/peter_zijlstra-frob-rcu.patch @@ -1,7 +1,7 @@ Subject: rcu: Frob softirq test From: Peter Zijlstra <a.p.zijlstra@chello.nl> Date: Sat Aug 13 00:23:17 CEST 2011 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz With RT_FULL we get the below wreckage: diff --git a/debian/patches/features/all/rt/peterz-srcu-crypto-chain.patch b/debian/patches/features/all/rt/peterz-srcu-crypto-chain.patch index 31cf19a199cf..6b58e8e8b0b5 100644 --- a/debian/patches/features/all/rt/peterz-srcu-crypto-chain.patch +++ b/debian/patches/features/all/rt/peterz-srcu-crypto-chain.patch @@ -1,7 +1,7 @@ Subject: crypto: Convert crypto notifier chain to SRCU From: Peter Zijlstra <peterz@infradead.org> Date: Fri, 05 Oct 2012 09:03:24 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The crypto notifier deadlocks on RT. Though this can be a real deadlock on mainline as well due to fifo fair rwsems. diff --git a/debian/patches/features/all/rt/pid.h-include-atomic.h.patch b/debian/patches/features/all/rt/pid.h-include-atomic.h.patch index 342d3ab8e679..f725136d03cc 100644 --- a/debian/patches/features/all/rt/pid.h-include-atomic.h.patch +++ b/debian/patches/features/all/rt/pid.h-include-atomic.h.patch @@ -1,7 +1,7 @@ From: Grygorii Strashko <Grygorii.Strashko@linaro.org> Date: Tue, 21 Jul 2015 19:43:56 +0300 Subject: wait.h: include atomic.h -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz This patch fixes build error: CC kernel/pid_namespace.o diff --git a/debian/patches/features/all/rt/ping-sysrq.patch b/debian/patches/features/all/rt/ping-sysrq.patch index e64c52d3251e..23b7afaaedf9 100644 --- a/debian/patches/features/all/rt/ping-sysrq.patch +++ b/debian/patches/features/all/rt/ping-sysrq.patch @@ -1,7 +1,7 @@ Subject: net: sysrq via icmp From: Carsten Emde <C.Emde@osadl.org> Date: Tue, 19 Jul 2011 13:51:17 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz There are (probably rare) situations when a system crashed and the system console becomes unresponsive but the network icmp layer still is alive. diff --git a/debian/patches/features/all/rt/posix-timers-no-broadcast.patch b/debian/patches/features/all/rt/posix-timers-no-broadcast.patch index 3a7b1c4b4913..6b2b955f75d7 100644 --- a/debian/patches/features/all/rt/posix-timers-no-broadcast.patch +++ b/debian/patches/features/all/rt/posix-timers-no-broadcast.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 3 Jul 2009 08:29:20 -0500 Subject: posix-timers: Prevent broadcast signals -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Posix timers should not send broadcast signals and kernel only signals. Prevent it. diff --git a/debian/patches/features/all/rt/posix-timers-thread-posix-cpu-timers-on-rt.patch b/debian/patches/features/all/rt/posix-timers-thread-posix-cpu-timers-on-rt.patch index fddebe388631..13c12b437b3d 100644 --- a/debian/patches/features/all/rt/posix-timers-thread-posix-cpu-timers-on-rt.patch +++ b/debian/patches/features/all/rt/posix-timers-thread-posix-cpu-timers-on-rt.patch @@ -1,7 +1,7 @@ From: John Stultz <johnstul@us.ibm.com> Date: Fri, 3 Jul 2009 08:29:58 -0500 Subject: posix-timers: Thread posix-cpu-timers on -rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz posix-cpu-timer code takes non -rt safe locks in hard irq context. Move it to a thread. @@ -43,7 +43,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> [PIDTYPE_PGID] = INIT_PID_LINK(PIDTYPE_PGID), \ --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -1542,6 +1542,9 @@ struct task_struct { +@@ -1543,6 +1543,9 @@ struct task_struct { struct task_cputime cputime_expires; struct list_head cpu_timers[3]; diff --git a/debian/patches/features/all/rt/power-disable-highmem-on-rt.patch b/debian/patches/features/all/rt/power-disable-highmem-on-rt.patch index 9a4a552e3fd6..21e7e630baa8 100644 --- a/debian/patches/features/all/rt/power-disable-highmem-on-rt.patch +++ b/debian/patches/features/all/rt/power-disable-highmem-on-rt.patch @@ -1,7 +1,7 @@ Subject: powerpc: Disable highmem on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 18 Jul 2011 17:08:34 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The current highmem handling on -RT is not compatible and needs fixups. diff --git a/debian/patches/features/all/rt/power-use-generic-rwsem-on-rt.patch b/debian/patches/features/all/rt/power-use-generic-rwsem-on-rt.patch index a75087c17e82..9add8f494ee9 100644 --- a/debian/patches/features/all/rt/power-use-generic-rwsem-on-rt.patch +++ b/debian/patches/features/all/rt/power-use-generic-rwsem-on-rt.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 14 Jul 2015 14:26:34 +0200 Subject: powerpc: Use generic rwsem on RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use generic code which uses rtmutex diff --git a/debian/patches/features/all/rt/powerpc-kvm-Disable-in-kernel-MPIC-emulation-for-PRE.patch b/debian/patches/features/all/rt/powerpc-kvm-Disable-in-kernel-MPIC-emulation-for-PRE.patch index 64533c0ae56a..46fc578277bc 100644 --- a/debian/patches/features/all/rt/powerpc-kvm-Disable-in-kernel-MPIC-emulation-for-PRE.patch +++ b/debian/patches/features/all/rt/powerpc-kvm-Disable-in-kernel-MPIC-emulation-for-PRE.patch @@ -1,7 +1,7 @@ From: Bogdan Purcareata <bogdan.purcareata@freescale.com> Date: Fri, 24 Apr 2015 15:53:13 +0000 Subject: powerpc/kvm: Disable in-kernel MPIC emulation for PREEMPT_RT_FULL -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz While converting the openpic emulation code to use a raw_spinlock_t enables guests to run on RT, there's still a performance issue. For interrupts sent in diff --git a/debian/patches/features/all/rt/powerpc-preempt-lazy-support.patch b/debian/patches/features/all/rt/powerpc-preempt-lazy-support.patch index fe77c005985e..2f7c1bad445a 100644 --- a/debian/patches/features/all/rt/powerpc-preempt-lazy-support.patch +++ b/debian/patches/features/all/rt/powerpc-preempt-lazy-support.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 1 Nov 2012 10:14:11 +0100 Subject: powerpc: Add support for lazy preemption -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Implement the powerpc pieces for lazy preempt. diff --git a/debian/patches/features/all/rt/powerpc-ps3-device-init.c-adapt-to-completions-using.patch b/debian/patches/features/all/rt/powerpc-ps3-device-init.c-adapt-to-completions-using.patch index f6cd6ed8236e..317b7e0d36ae 100644 --- a/debian/patches/features/all/rt/powerpc-ps3-device-init.c-adapt-to-completions-using.patch +++ b/debian/patches/features/all/rt/powerpc-ps3-device-init.c-adapt-to-completions-using.patch @@ -1,7 +1,7 @@ From: Paul Gortmaker <paul.gortmaker@windriver.com> Date: Sun, 31 May 2015 14:44:42 -0400 Subject: powerpc: ps3/device-init.c - adapt to completions using swait vs wait -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz To fix: diff --git a/debian/patches/features/all/rt/preempt-lazy-support.patch b/debian/patches/features/all/rt/preempt-lazy-support.patch index 05c24a1c1fb8..5d469c680736 100644 --- a/debian/patches/features/all/rt/preempt-lazy-support.patch +++ b/debian/patches/features/all/rt/preempt-lazy-support.patch @@ -1,7 +1,7 @@ Subject: sched: Add support for lazy preemption From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 26 Oct 2012 18:50:54 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz It has become an obsession to mitigate the determinism vs. throughput loss of RT. Looking at the mainline semantics of preemption points @@ -166,7 +166,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -2961,6 +2961,43 @@ static inline int test_tsk_need_resched( +@@ -2963,6 +2963,43 @@ static inline int test_tsk_need_resched( return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED)); } diff --git a/debian/patches/features/all/rt/preempt-nort-rt-variants.patch b/debian/patches/features/all/rt/preempt-nort-rt-variants.patch index db0434eb41ff..e236c9953bcf 100644 --- a/debian/patches/features/all/rt/preempt-nort-rt-variants.patch +++ b/debian/patches/features/all/rt/preempt-nort-rt-variants.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 24 Jul 2009 12:38:56 +0200 Subject: preempt: Provide preempt_*_(no)rt variants -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz RT needs a few preempt_disable/enable points which are not necessary otherwise. Implement variants to avoid #ifdeffery. diff --git a/debian/patches/features/all/rt/printk-27force_early_printk-27-boot-param-to-help-with-debugging.patch b/debian/patches/features/all/rt/printk-27force_early_printk-27-boot-param-to-help-with-debugging.patch index 503d2025815d..f4ff3548ff91 100644 --- a/debian/patches/features/all/rt/printk-27force_early_printk-27-boot-param-to-help-with-debugging.patch +++ b/debian/patches/features/all/rt/printk-27force_early_printk-27-boot-param-to-help-with-debugging.patch @@ -1,7 +1,7 @@ Subject: printk: Add "force_early_printk" boot param to help with debugging From: Peter Zijlstra <peterz@infradead.org> Date: Fri, 02 Sep 2011 14:41:29 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Gives me an option to screw printk and actually see what the machine says. diff --git a/debian/patches/features/all/rt/printk-kill.patch b/debian/patches/features/all/rt/printk-kill.patch index 9b7482be74aa..a146684b0a0f 100644 --- a/debian/patches/features/all/rt/printk-kill.patch +++ b/debian/patches/features/all/rt/printk-kill.patch @@ -1,7 +1,7 @@ Subject: printk: Add a printk kill switch From: Ingo Molnar <mingo@elte.hu> Date: Fri, 22 Jul 2011 17:58:40 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Add a prinkt-kill-switch. This is used from (NMI) watchdog to ensure that it does not dead-lock with the early printk code. diff --git a/debian/patches/features/all/rt/printk-rt-aware.patch b/debian/patches/features/all/rt/printk-rt-aware.patch index 990d6049a870..2ea38627dd2e 100644 --- a/debian/patches/features/all/rt/printk-rt-aware.patch +++ b/debian/patches/features/all/rt/printk-rt-aware.patch @@ -1,7 +1,7 @@ Subject: printk: Make rt aware From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 19 Sep 2012 14:50:37 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Drop the lock before calling the console driver and do not disable interrupts while printing to a serial console. diff --git a/debian/patches/features/all/rt/ptrace-fix-ptrace-vs-tasklist_lock-race.patch b/debian/patches/features/all/rt/ptrace-fix-ptrace-vs-tasklist_lock-race.patch index 1e54f4d922f0..b3cbba8d88ce 100644 --- a/debian/patches/features/all/rt/ptrace-fix-ptrace-vs-tasklist_lock-race.patch +++ b/debian/patches/features/all/rt/ptrace-fix-ptrace-vs-tasklist_lock-race.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Thu, 29 Aug 2013 18:21:04 +0200 Subject: ptrace: fix ptrace vs tasklist_lock race -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz As explained by Alexander Fyodorov <halcy@yandex.ru>: @@ -42,7 +42,7 @@ Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> #define task_contributes_to_load(task) \ ((task->state & TASK_UNINTERRUPTIBLE) != 0 && \ (task->flags & PF_FROZEN) == 0 && \ -@@ -2981,6 +2978,51 @@ static inline int signal_pending_state(l +@@ -2983,6 +2980,51 @@ static inline int signal_pending_state(l return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p); } diff --git a/debian/patches/features/all/rt/radix-tree-rt-aware.patch b/debian/patches/features/all/rt/radix-tree-rt-aware.patch index 4f440d1ec495..1542e32cf765 100644 --- a/debian/patches/features/all/rt/radix-tree-rt-aware.patch +++ b/debian/patches/features/all/rt/radix-tree-rt-aware.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 21:33:18 +0200 Subject: radix-tree: Make RT aware -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Disable radix_tree_preload() on -RT. This functions returns with preemption disabled which may cause high latencies and breaks if the diff --git a/debian/patches/features/all/rt/random-make-it-work-on-rt.patch b/debian/patches/features/all/rt/random-make-it-work-on-rt.patch index 8c71e6d8c94a..b4860c04a83b 100644 --- a/debian/patches/features/all/rt/random-make-it-work-on-rt.patch +++ b/debian/patches/features/all/rt/random-make-it-work-on-rt.patch @@ -1,7 +1,7 @@ Subject: random: Make it work on rt From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 21 Aug 2012 20:38:50 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Delegate the random insertion to the forced threaded interrupt handler. Store the return IP of the hard interrupt handler in the irq diff --git a/debian/patches/features/all/rt/rcu-Eliminate-softirq-processing-from-rcutree.patch b/debian/patches/features/all/rt/rcu-Eliminate-softirq-processing-from-rcutree.patch index 21f6e2f96866..e8abd0244407 100644 --- a/debian/patches/features/all/rt/rcu-Eliminate-softirq-processing-from-rcutree.patch +++ b/debian/patches/features/all/rt/rcu-Eliminate-softirq-processing-from-rcutree.patch @@ -1,7 +1,7 @@ From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> Date: Mon, 4 Nov 2013 13:21:10 -0800 Subject: rcu: Eliminate softirq processing from rcutree -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Running RCU out of softirq is a problem for some workloads that would like to manage RCU core processing independently of other softirq work, diff --git a/debian/patches/features/all/rt/rcu-disable-rcu-fast-no-hz-on-rt.patch b/debian/patches/features/all/rt/rcu-disable-rcu-fast-no-hz-on-rt.patch index 4bf4b4f4f356..27359fe92c75 100644 --- a/debian/patches/features/all/rt/rcu-disable-rcu-fast-no-hz-on-rt.patch +++ b/debian/patches/features/all/rt/rcu-disable-rcu-fast-no-hz-on-rt.patch @@ -1,7 +1,7 @@ Subject: rcu: Disable RCU_FAST_NO_HZ on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 28 Oct 2012 13:26:09 +0000 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz This uses a timer_list timer from the irq disabled guts of the idle code. Disable it for now to prevent wreckage. diff --git a/debian/patches/features/all/rt/rcu-make-RCU_BOOST-default-on-RT.patch b/debian/patches/features/all/rt/rcu-make-RCU_BOOST-default-on-RT.patch index b94e1f8efb4f..b5edca7010f3 100644 --- a/debian/patches/features/all/rt/rcu-make-RCU_BOOST-default-on-RT.patch +++ b/debian/patches/features/all/rt/rcu-make-RCU_BOOST-default-on-RT.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 21 Mar 2014 20:19:05 +0100 Subject: rcu: make RCU_BOOST default on RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Since it is no longer invoked from the softirq people run into OOM more often if the priority of the RCU thread is too low. Making boosting diff --git a/debian/patches/features/all/rt/rcu-merge-rcu-bh-into-rcu-preempt-for-rt.patch b/debian/patches/features/all/rt/rcu-merge-rcu-bh-into-rcu-preempt-for-rt.patch index bffedb3cf097..905495baae57 100644 --- a/debian/patches/features/all/rt/rcu-merge-rcu-bh-into-rcu-preempt-for-rt.patch +++ b/debian/patches/features/all/rt/rcu-merge-rcu-bh-into-rcu-preempt-for-rt.patch @@ -1,7 +1,7 @@ Subject: rcu: Merge RCU-bh into RCU-preempt Date: Wed, 5 Oct 2011 11:59:38 -0700 From: Thomas Gleixner <tglx@linutronix.de> -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The Linux kernel has long RCU-bh read-side critical sections that intolerably increase scheduling latency under mainline's RCU-bh rules, diff --git a/debian/patches/features/all/rt/rcu-more-swait-conversions.patch b/debian/patches/features/all/rt/rcu-more-swait-conversions.patch index 2c4d95df131e..85adf0f1b531 100644 --- a/debian/patches/features/all/rt/rcu-more-swait-conversions.patch +++ b/debian/patches/features/all/rt/rcu-more-swait-conversions.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 31 Jul 2013 19:00:35 +0200 Subject: rcu: use simple waitqueues -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Convert RCU's wait-queues into simple waitqueues. diff --git a/debian/patches/features/all/rt/rcutree-rcu_bh_qs-disable-irq-while-calling-rcu_pree.patch b/debian/patches/features/all/rt/rcutree-rcu_bh_qs-disable-irq-while-calling-rcu_pree.patch index 52be9161758a..91e05b76df29 100644 --- a/debian/patches/features/all/rt/rcutree-rcu_bh_qs-disable-irq-while-calling-rcu_pree.patch +++ b/debian/patches/features/all/rt/rcutree-rcu_bh_qs-disable-irq-while-calling-rcu_pree.patch @@ -1,7 +1,7 @@ From: Tiejun Chen <tiejun.chen@windriver.com> Date: Wed, 18 Dec 2013 17:51:49 +0800 Subject: rcutree/rcu_bh_qs: Disable irq while calling rcu_preempt_qs() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Any callers to the function rcu_preempt_qs() must disable irqs in order to protect the assignment to ->rcu_read_unlock_special. In diff --git a/debian/patches/features/all/rt/re-migrate_disable-race-with-cpu-hotplug-3f.patch b/debian/patches/features/all/rt/re-migrate_disable-race-with-cpu-hotplug-3f.patch index 3b9248929a0f..b3130384b3fb 100644 --- a/debian/patches/features/all/rt/re-migrate_disable-race-with-cpu-hotplug-3f.patch +++ b/debian/patches/features/all/rt/re-migrate_disable-race-with-cpu-hotplug-3f.patch @@ -1,7 +1,7 @@ From: Yong Zhang <yong.zhang0@gmail.com> Date: Thu, 28 Jul 2011 11:16:00 +0800 Subject: hotplug: Reread hotplug_pcp on pin_current_cpu() retry -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz When retry happens, it's likely that the task has been migrated to another cpu (except unplug failed), but it still derefernces the diff --git a/debian/patches/features/all/rt/re-preempt_rt_full-arm-coredump-fails-for-cpu-3e-3d-4.patch b/debian/patches/features/all/rt/re-preempt_rt_full-arm-coredump-fails-for-cpu-3e-3d-4.patch index 880ef1dd744b..830db4d7cadd 100644 --- a/debian/patches/features/all/rt/re-preempt_rt_full-arm-coredump-fails-for-cpu-3e-3d-4.patch +++ b/debian/patches/features/all/rt/re-preempt_rt_full-arm-coredump-fails-for-cpu-3e-3d-4.patch @@ -1,7 +1,7 @@ Subject: ARM: Initialize split page table locks for vector page From: Frank Rowand <frank.rowand@am.sony.com> Date: Sat, 1 Oct 2011 18:58:13 -0700 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Without this patch, ARM can not use SPLIT_PTLOCK_CPUS if PREEMPT_RT_FULL=y because vectors_user_mapping() creates a diff --git a/debian/patches/features/all/rt/relay-fix-timer-madness.patch b/debian/patches/features/all/rt/relay-fix-timer-madness.patch index 5e9ae068d4b9..e741f64e8600 100644 --- a/debian/patches/features/all/rt/relay-fix-timer-madness.patch +++ b/debian/patches/features/all/rt/relay-fix-timer-madness.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:44:07 -0500 Subject: relay: Fix timer madness -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz remove timer calls (!!!) from deep within the tracing infrastructure. This was totally bogus code that can cause lockups and worse. Poll diff --git a/debian/patches/features/all/rt/rfc-arm-smp-__cpu_disable-fix-sleeping-function-called-from-invalid-context.patch b/debian/patches/features/all/rt/rfc-arm-smp-__cpu_disable-fix-sleeping-function-called-from-invalid-context.patch index 5ff01bac792f..98f946bc2c07 100644 --- a/debian/patches/features/all/rt/rfc-arm-smp-__cpu_disable-fix-sleeping-function-called-from-invalid-context.patch +++ b/debian/patches/features/all/rt/rfc-arm-smp-__cpu_disable-fix-sleeping-function-called-from-invalid-context.patch @@ -1,7 +1,7 @@ Subject: ARM: smp: Move clear_tasks_mm_cpumask() call to __cpu_die() From: Grygorii Strashko <grygorii.strashko@ti.com> Date: Fri, 11 Sep 2015 21:21:23 +0300 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz When running with the RT-kernel (4.1.5-rt5) on TI OMAP dra7-evm and trying to do Suspend to RAM, the following backtrace occurs: diff --git a/debian/patches/features/all/rt/rt-add-rt-locks.patch b/debian/patches/features/all/rt/rt-add-rt-locks.patch index 900af2cd392a..b4dfad1884dc 100644 --- a/debian/patches/features/all/rt/rt-add-rt-locks.patch +++ b/debian/patches/features/all/rt/rt-add-rt-locks.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 26 Jul 2009 19:39:56 +0200 Subject: rt: Add the preempt-rt lock replacement APIs -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Map spinlocks, rwlocks, rw_semaphores and semaphores to the rt_mutex based locking functions for preempt-rt. diff --git a/debian/patches/features/all/rt/rt-introduce-cpu-chill.patch b/debian/patches/features/all/rt/rt-introduce-cpu-chill.patch index 91f2db1330a8..ff98cadea652 100644 --- a/debian/patches/features/all/rt/rt-introduce-cpu-chill.patch +++ b/debian/patches/features/all/rt/rt-introduce-cpu-chill.patch @@ -1,7 +1,7 @@ Subject: rt: Introduce cpu_chill() From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 07 Mar 2012 20:51:03 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Retry loops on RT might loop forever when the modifying side was preempted. Add cpu_chill() to replace cpu_relax(). cpu_chill() diff --git a/debian/patches/features/all/rt/rt-local-irq-lock.patch b/debian/patches/features/all/rt/rt-local-irq-lock.patch index b74878c45db7..ca27478d925e 100644 --- a/debian/patches/features/all/rt/rt-local-irq-lock.patch +++ b/debian/patches/features/all/rt/rt-local-irq-lock.patch @@ -1,7 +1,7 @@ Subject: rt: Add local irq locks From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 20 Jun 2011 09:03:47 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Introduce locallock. For !RT this maps to preempt_disable()/ local_irq_disable() so there is not much that changes. For RT this will diff --git a/debian/patches/features/all/rt/rt-preempt-base-config.patch b/debian/patches/features/all/rt/rt-preempt-base-config.patch index b4cd4aec1f0b..8f8a9d4309db 100644 --- a/debian/patches/features/all/rt/rt-preempt-base-config.patch +++ b/debian/patches/features/all/rt/rt-preempt-base-config.patch @@ -1,7 +1,7 @@ Subject: rt: Provide PREEMPT_RT_BASE config switch From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 17 Jun 2011 12:39:57 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Introduce PREEMPT_RT_BASE which enables parts of PREEMPT_RT_FULL. Forces interrupt threading and enables some of the RT diff --git a/debian/patches/features/all/rt/rt-serial-warn-fix.patch b/debian/patches/features/all/rt/rt-serial-warn-fix.patch index dee4d0e9bf2e..4ffae1059d49 100644 --- a/debian/patches/features/all/rt/rt-serial-warn-fix.patch +++ b/debian/patches/features/all/rt/rt-serial-warn-fix.patch @@ -1,7 +1,7 @@ Subject: rt: Improve the serial console PASS_LIMIT From: Ingo Molnar <mingo@elte.hu> Date: Wed Dec 14 13:05:54 CET 2011 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Beyond the warning: diff --git a/debian/patches/features/all/rt/rtmutex--Handle-non-enqueued-waiters-gracefully b/debian/patches/features/all/rt/rtmutex--Handle-non-enqueued-waiters-gracefully index ee6bfad5b06e..6926b68e0245 100644 --- a/debian/patches/features/all/rt/rtmutex--Handle-non-enqueued-waiters-gracefully +++ b/debian/patches/features/all/rt/rtmutex--Handle-non-enqueued-waiters-gracefully @@ -1,7 +1,7 @@ Subject: rtmutex: Handle non enqueued waiters gracefully From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 06 Nov 2015 18:51:03 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Yimin debugged that in case of a PI wakeup in progress when rt_mutex_start_proxy_lock() calls task_blocks_on_rt_mutex() the latter diff --git a/debian/patches/features/all/rt/rtmutex-Use-chainwalking-control-enum.patch b/debian/patches/features/all/rt/rtmutex-Use-chainwalking-control-enum.patch index eb9597228c16..06a14a597ab8 100644 --- a/debian/patches/features/all/rt/rtmutex-Use-chainwalking-control-enum.patch +++ b/debian/patches/features/all/rt/rtmutex-Use-chainwalking-control-enum.patch @@ -2,7 +2,7 @@ From 13f032043086194982ac91c68124adae545f5627 Mon Sep 17 00:00:00 2001 From: "bmouring@ni.com" <bmouring@ni.com> Date: Tue, 15 Dec 2015 17:07:30 -0600 Subject: [PATCH] rtmutex: Use chainwalking control enum -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz In 8930ed80 (rtmutex: Cleanup deadlock detector debug logic), chainwalking control enums were introduced to limit the deadlock diff --git a/debian/patches/features/all/rt/rtmutex-add-a-first-shot-of-ww_mutex.patch b/debian/patches/features/all/rt/rtmutex-add-a-first-shot-of-ww_mutex.patch index 7143ed49cdf9..d47c90e40457 100644 --- a/debian/patches/features/all/rt/rtmutex-add-a-first-shot-of-ww_mutex.patch +++ b/debian/patches/features/all/rt/rtmutex-add-a-first-shot-of-ww_mutex.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <sebastian@breakpoint.cc> Date: Mon, 28 Oct 2013 09:36:37 +0100 Subject: rtmutex: Add RT aware ww locks -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz lockdep says: | -------------------------------------------------------------------------- diff --git a/debian/patches/features/all/rt/rtmutex-avoid-include-hell.patch b/debian/patches/features/all/rt/rtmutex-avoid-include-hell.patch index 1ccb6b6e5765..539823011730 100644 --- a/debian/patches/features/all/rt/rtmutex-avoid-include-hell.patch +++ b/debian/patches/features/all/rt/rtmutex-avoid-include-hell.patch @@ -1,7 +1,7 @@ Subject: rtmutex: Avoid include hell From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 29 Jun 2011 20:06:39 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Include only the required raw types. This avoids pulling in the complete spinlock header which in turn requires rtmutex.h at some point. diff --git a/debian/patches/features/all/rt/rtmutex-futex-prepare-rt.patch b/debian/patches/features/all/rt/rtmutex-futex-prepare-rt.patch index e88ebe200893..9364219851c9 100644 --- a/debian/patches/features/all/rt/rtmutex-futex-prepare-rt.patch +++ b/debian/patches/features/all/rt/rtmutex-futex-prepare-rt.patch @@ -1,7 +1,7 @@ Subject: rtmutex: Handle the various new futex race conditions From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 10 Jun 2011 11:04:15 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz RT opens a few new interesting race conditions in the rtmutex/futex combo due to futex hash bucket lock being a 'sleeping' spinlock and diff --git a/debian/patches/features/all/rt/rtmutex-lock-killable.patch b/debian/patches/features/all/rt/rtmutex-lock-killable.patch index 1cd97a205cd0..5873a6655182 100644 --- a/debian/patches/features/all/rt/rtmutex-lock-killable.patch +++ b/debian/patches/features/all/rt/rtmutex-lock-killable.patch @@ -1,7 +1,7 @@ Subject: rtmutex: Add rtmutex_lock_killable() From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 09 Jun 2011 11:43:52 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Add "killable" type to rtmutex. We need this since rtmutex are used as "normal" mutexes which do use this type. diff --git a/debian/patches/features/all/rt/rtmutex-trylock-is-okay-on-RT.patch b/debian/patches/features/all/rt/rtmutex-trylock-is-okay-on-RT.patch index 92903347294e..b36988a6c769 100644 --- a/debian/patches/features/all/rt/rtmutex-trylock-is-okay-on-RT.patch +++ b/debian/patches/features/all/rt/rtmutex-trylock-is-okay-on-RT.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed 02 Dec 2015 11:34:07 +0100 Subject: rtmutex: trylock is okay on -RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz non-RT kernel could deadlock on rt_mutex_trylock() in softirq context. On -RT we don't run softirqs in IRQ context but in thread context so it is diff --git a/debian/patches/features/all/rt/rtmutex_dont_include_rcu.patch b/debian/patches/features/all/rt/rtmutex_dont_include_rcu.patch index 032a5dab6c45..d99099de11a1 100644 --- a/debian/patches/features/all/rt/rtmutex_dont_include_rcu.patch +++ b/debian/patches/features/all/rt/rtmutex_dont_include_rcu.patch @@ -1,6 +1,6 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Subject: rbtree: don't include the rcu header -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The RCU header pulls in spinlock.h and fails due not yet defined types: diff --git a/debian/patches/features/all/rt/sas-ata-isci-dont-t-disable-interrupts-in-qc_issue-h.patch b/debian/patches/features/all/rt/sas-ata-isci-dont-t-disable-interrupts-in-qc_issue-h.patch index 230236909302..985570c5c2d5 100644 --- a/debian/patches/features/all/rt/sas-ata-isci-dont-t-disable-interrupts-in-qc_issue-h.patch +++ b/debian/patches/features/all/rt/sas-ata-isci-dont-t-disable-interrupts-in-qc_issue-h.patch @@ -1,7 +1,7 @@ From: Paul Gortmaker <paul.gortmaker@windriver.com> Date: Sat, 14 Feb 2015 11:01:16 -0500 Subject: sas-ata/isci: dont't disable interrupts in qc_issue handler -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On 3.14-rt we see the following trace on Canoe Pass for SCSI_ISCI "Intel(R) C600 Series Chipset SAS Controller" diff --git a/debian/patches/features/all/rt/sched-deadline-dl_task_timer-has-to-be-irqsafe.patch b/debian/patches/features/all/rt/sched-deadline-dl_task_timer-has-to-be-irqsafe.patch index b188e3b17d2d..2e7b024a43a6 100644 --- a/debian/patches/features/all/rt/sched-deadline-dl_task_timer-has-to-be-irqsafe.patch +++ b/debian/patches/features/all/rt/sched-deadline-dl_task_timer-has-to-be-irqsafe.patch @@ -1,7 +1,7 @@ From: Juri Lelli <juri.lelli@gmail.com> Date: Tue, 13 May 2014 15:30:20 +0200 Subject: sched/deadline: dl_task_timer has to be irqsafe -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz As for rt_period_timer, dl_task_timer has to be irqsafe. diff --git a/debian/patches/features/all/rt/sched-delay-put-task.patch b/debian/patches/features/all/rt/sched-delay-put-task.patch index 85db90b03fc6..3bd2e453e79f 100644 --- a/debian/patches/features/all/rt/sched-delay-put-task.patch +++ b/debian/patches/features/all/rt/sched-delay-put-task.patch @@ -1,7 +1,7 @@ Subject: sched: Move task_struct cleanup to RCU From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 31 May 2011 16:59:16 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz __put_task_struct() does quite some expensive work. We don't want to burden random tasks with that. @@ -14,7 +14,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -1829,6 +1829,9 @@ struct task_struct { +@@ -1830,6 +1830,9 @@ struct task_struct { unsigned int sequential_io; unsigned int sequential_io_avg; #endif @@ -24,7 +24,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> #ifdef CONFIG_DEBUG_ATOMIC_SLEEP unsigned long task_state_change; #endif -@@ -2037,6 +2040,15 @@ extern struct pid *cad_pid; +@@ -2039,6 +2042,15 @@ extern struct pid *cad_pid; extern void free_task(struct task_struct *tsk); #define get_task_struct(tsk) do { atomic_inc(&(tsk)->usage); } while(0) @@ -40,7 +40,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> extern void __put_task_struct(struct task_struct *t); static inline void put_task_struct(struct task_struct *t) -@@ -2044,6 +2056,7 @@ static inline void put_task_struct(struc +@@ -2046,6 +2058,7 @@ static inline void put_task_struct(struc if (atomic_dec_and_test(&t->usage)) __put_task_struct(t); } diff --git a/debian/patches/features/all/rt/sched-disable-rt-group-sched-on-rt.patch b/debian/patches/features/all/rt/sched-disable-rt-group-sched-on-rt.patch index e01102d0f826..c7df8502fb00 100644 --- a/debian/patches/features/all/rt/sched-disable-rt-group-sched-on-rt.patch +++ b/debian/patches/features/all/rt/sched-disable-rt-group-sched-on-rt.patch @@ -1,7 +1,7 @@ Subject: sched: Disable CONFIG_RT_GROUP_SCHED on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 18 Jul 2011 17:03:52 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Carsten reported problems when running: diff --git a/debian/patches/features/all/rt/sched-disable-ttwu-queue.patch b/debian/patches/features/all/rt/sched-disable-ttwu-queue.patch index 8f67861904d1..7cb92c02adf9 100644 --- a/debian/patches/features/all/rt/sched-disable-ttwu-queue.patch +++ b/debian/patches/features/all/rt/sched-disable-ttwu-queue.patch @@ -1,7 +1,7 @@ Subject: sched: Disable TTWU_QUEUE on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 13 Sep 2011 16:42:35 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The queued remote wakeup mechanism can introduce rather large latencies if the number of migrated tasks is high. Disable it for RT. diff --git a/debian/patches/features/all/rt/sched-limit-nr-migrate.patch b/debian/patches/features/all/rt/sched-limit-nr-migrate.patch index cb137bd252c1..cc55e81dcbfb 100644 --- a/debian/patches/features/all/rt/sched-limit-nr-migrate.patch +++ b/debian/patches/features/all/rt/sched-limit-nr-migrate.patch @@ -1,7 +1,7 @@ Subject: sched: Limit the number of task migrations per batch From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 06 Jun 2011 12:12:51 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Put an upper limit on the number of tasks which are migrated per batch to avoid large latencies. diff --git a/debian/patches/features/all/rt/sched-might-sleep-do-not-account-rcu-depth.patch b/debian/patches/features/all/rt/sched-might-sleep-do-not-account-rcu-depth.patch index f0224c12f801..b1cd8f2eace6 100644 --- a/debian/patches/features/all/rt/sched-might-sleep-do-not-account-rcu-depth.patch +++ b/debian/patches/features/all/rt/sched-might-sleep-do-not-account-rcu-depth.patch @@ -1,7 +1,7 @@ Subject: sched: Do not account rcu_preempt_depth on RT in might_sleep() From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 07 Jun 2011 09:19:06 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz RT changes the rcu_preempt_depth semantics, so we cannot check for it in might_sleep(). diff --git a/debian/patches/features/all/rt/sched-mmdrop-delayed.patch b/debian/patches/features/all/rt/sched-mmdrop-delayed.patch index af2b08d13188..909d328c9dc6 100644 --- a/debian/patches/features/all/rt/sched-mmdrop-delayed.patch +++ b/debian/patches/features/all/rt/sched-mmdrop-delayed.patch @@ -1,7 +1,7 @@ Subject: sched: Move mmdrop to RCU on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 06 Jun 2011 12:20:33 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Takes sleeping locks and calls into the memory allocator, so nothing we want to do in task switch and oder atomic contexts. @@ -36,7 +36,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> void __user *bd_addr; --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -2603,12 +2603,24 @@ extern struct mm_struct * mm_alloc(void) +@@ -2605,12 +2605,24 @@ extern struct mm_struct * mm_alloc(void) /* mmdrop drops the mm and the page tables */ extern void __mmdrop(struct mm_struct *); diff --git a/debian/patches/features/all/rt/sched-reset-task-s-lockless-wake-queues-on-fork.patch b/debian/patches/features/all/rt/sched-reset-task-s-lockless-wake-queues-on-fork.patch deleted file mode 100644 index 29a1139cd679..000000000000 --- a/debian/patches/features/all/rt/sched-reset-task-s-lockless-wake-queues-on-fork.patch +++ /dev/null @@ -1,35 +0,0 @@ -From c4c38e7dcd4e925f624cc7fe18aeaad841fd7d6f Mon Sep 17 00:00:00 2001 -From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> -Date: Mon, 21 Dec 2015 18:17:10 +0100 -Subject: [PATCH] sched: reset task's lockless wake-queues on fork() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz - -In 7675104990ed ("sched: Implement lockless wake-queues") we gained -lockless wake-queues. -RT managed to lockup itself with those. There -could be multiple attempts for task X to enqueue it for a wakeup -_even_ if task X is already running. -The reason is that task X could be runnable but not yet on CPU. The the -task performing the wakeup did not leave the CPU it could performe -multiple wakeups. -With the proper timming task X could be running and enqueued for a -wakeup. If this happens while X is performing a fork() then its its -child will have a !NULL `wake_q` member copied. -This is not a problem as long as the child task does not participate in -lockless wakeups :) - -Fixes: 7675104990ed ("sched: Implement lockless wake-queues") -Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> ---- - kernel/fork.c | 1 + - 1 file changed, 1 insertion(+) - ---- a/kernel/fork.c -+++ b/kernel/fork.c -@@ -380,6 +380,7 @@ static struct task_struct *dup_task_stru - #endif - tsk->splice_pipe = NULL; - tsk->task_frag.page = NULL; -+ tsk->wake_q.next = NULL; - - account_kernel_stack(ti, 1); - diff --git a/debian/patches/features/all/rt/sched-rt-mutex-wakeup.patch b/debian/patches/features/all/rt/sched-rt-mutex-wakeup.patch index 48350deccf4d..058926917d1b 100644 --- a/debian/patches/features/all/rt/sched-rt-mutex-wakeup.patch +++ b/debian/patches/features/all/rt/sched-rt-mutex-wakeup.patch @@ -1,7 +1,7 @@ Subject: sched: Add saved_state for tasks blocked on sleeping locks From: Thomas Gleixner <tglx@linutronix.de> Date: Sat, 25 Jun 2011 09:21:04 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Spinlocks are state preserving in !RT. RT changes the state when a task gets blocked on a lock. So we need to remember the state before @@ -26,7 +26,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> void *stack; atomic_t usage; unsigned int flags; /* per process flags, defined below */ -@@ -2480,6 +2481,7 @@ extern void xtime_update(unsigned long t +@@ -2482,6 +2483,7 @@ extern void xtime_update(unsigned long t extern int wake_up_state(struct task_struct *tsk, unsigned int state); extern int wake_up_process(struct task_struct *tsk); diff --git a/debian/patches/features/all/rt/sched-ttwu-ensure-success-return-is-correct.patch b/debian/patches/features/all/rt/sched-ttwu-ensure-success-return-is-correct.patch index e7a12d5f3d45..ddec337c4fe9 100644 --- a/debian/patches/features/all/rt/sched-ttwu-ensure-success-return-is-correct.patch +++ b/debian/patches/features/all/rt/sched-ttwu-ensure-success-return-is-correct.patch @@ -1,7 +1,7 @@ Subject: sched: ttwu: Return success when only changing the saved_state value From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 13 Dec 2011 21:42:19 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz When a task blocks on a rt lock, it saves the current state in p->saved_state, so a lock related wake up will not destroy the diff --git a/debian/patches/features/all/rt/sched-workqueue-Only-wake-up-idle-workers-if-not-blo.patch b/debian/patches/features/all/rt/sched-workqueue-Only-wake-up-idle-workers-if-not-blo.patch index 9964dfc44e08..b38f81da5609 100644 --- a/debian/patches/features/all/rt/sched-workqueue-Only-wake-up-idle-workers-if-not-blo.patch +++ b/debian/patches/features/all/rt/sched-workqueue-Only-wake-up-idle-workers-if-not-blo.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Mon, 18 Mar 2013 15:12:49 -0400 Subject: sched/workqueue: Only wake up idle workers if not blocked on sleeping spin lock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz In -rt, most spin_locks() turn into mutexes. One of these spin_lock conversions is performed on the workqueue gcwq->lock. When the idle diff --git a/debian/patches/features/all/rt/scsi-fcoe-rt-aware.patch b/debian/patches/features/all/rt/scsi-fcoe-rt-aware.patch index 156a1486fc51..870be57de95d 100644 --- a/debian/patches/features/all/rt/scsi-fcoe-rt-aware.patch +++ b/debian/patches/features/all/rt/scsi-fcoe-rt-aware.patch @@ -1,7 +1,7 @@ Subject: scsi/fcoe: Make RT aware. From: Thomas Gleixner <tglx@linutronix.de> Date: Sat, 12 Nov 2011 14:00:48 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Do not disable preemption while taking sleeping locks. All user look safe for migrate_diable() only. diff --git a/debian/patches/features/all/rt/scsi-qla2xxx-fix-bug-sleeping-function-called-from-invalid-context.patch b/debian/patches/features/all/rt/scsi-qla2xxx-fix-bug-sleeping-function-called-from-invalid-context.patch index 0f0c498eff7d..8aa2fdc0de0a 100644 --- a/debian/patches/features/all/rt/scsi-qla2xxx-fix-bug-sleeping-function-called-from-invalid-context.patch +++ b/debian/patches/features/all/rt/scsi-qla2xxx-fix-bug-sleeping-function-called-from-invalid-context.patch @@ -1,7 +1,7 @@ Subject: scsi: qla2xxx: Use local_irq_save_nort() in qla2x00_poll From: John Kacur <jkacur@redhat.com> Date: Fri, 27 Apr 2012 12:48:46 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz RT triggers the following: diff --git a/debian/patches/features/all/rt/seqlock-prevent-rt-starvation.patch b/debian/patches/features/all/rt/seqlock-prevent-rt-starvation.patch index 70fc20e44d13..f4adbd112892 100644 --- a/debian/patches/features/all/rt/seqlock-prevent-rt-starvation.patch +++ b/debian/patches/features/all/rt/seqlock-prevent-rt-starvation.patch @@ -1,7 +1,7 @@ Subject: seqlock: Prevent rt starvation From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 22 Feb 2012 12:03:30 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz If a low prio writer gets preempted while holding the seqlock write locked, a high prio reader spins forever on RT. diff --git a/debian/patches/features/all/rt/signal-fix-up-rcu-wreckage.patch b/debian/patches/features/all/rt/signal-fix-up-rcu-wreckage.patch index 175f76e4e952..24e1ef502c92 100644 --- a/debian/patches/features/all/rt/signal-fix-up-rcu-wreckage.patch +++ b/debian/patches/features/all/rt/signal-fix-up-rcu-wreckage.patch @@ -1,7 +1,7 @@ Subject: signal: Make __lock_task_sighand() RT aware From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 22 Jul 2011 08:07:08 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz local_irq_save() + spin_lock(&sighand->siglock) does not work on -RT. Use the nort variants. diff --git a/debian/patches/features/all/rt/signal-revert-ptrace-preempt-magic.patch b/debian/patches/features/all/rt/signal-revert-ptrace-preempt-magic.patch index 503bbc480f27..96dd34e672c1 100644 --- a/debian/patches/features/all/rt/signal-revert-ptrace-preempt-magic.patch +++ b/debian/patches/features/all/rt/signal-revert-ptrace-preempt-magic.patch @@ -1,7 +1,7 @@ Subject: signal: Revert ptrace preempt magic From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 21 Sep 2011 19:57:12 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Upstream commit '53da1d9456fe7f8 fix ptrace slowness' is nothing more than a bandaid around the ptrace design trainwreck. It's not a diff --git a/debian/patches/features/all/rt/signals-allow-rt-tasks-to-cache-one-sigqueue-struct.patch b/debian/patches/features/all/rt/signals-allow-rt-tasks-to-cache-one-sigqueue-struct.patch index 8ce7e713d859..8f4132f9d910 100644 --- a/debian/patches/features/all/rt/signals-allow-rt-tasks-to-cache-one-sigqueue-struct.patch +++ b/debian/patches/features/all/rt/signals-allow-rt-tasks-to-cache-one-sigqueue-struct.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 3 Jul 2009 08:44:56 -0500 Subject: signals: Allow rt tasks to cache one sigqueue struct -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz To avoid allocation allow rt tasks to cache one sigqueue struct in task struct. @@ -18,7 +18,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -1566,6 +1566,7 @@ struct task_struct { +@@ -1567,6 +1567,7 @@ struct task_struct { /* signal handlers */ struct signal_struct *signal; struct sighand_struct *sighand; diff --git a/debian/patches/features/all/rt/skbufhead-raw-lock.patch b/debian/patches/features/all/rt/skbufhead-raw-lock.patch index 9859a9cb20da..5981139c9ef8 100644 --- a/debian/patches/features/all/rt/skbufhead-raw-lock.patch +++ b/debian/patches/features/all/rt/skbufhead-raw-lock.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 12 Jul 2011 15:38:34 +0200 Subject: net: Use skbufhead with raw lock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use the rps lock as rawlock so we can keep irq-off regions. It looks low latency. However we can't kfree() from this context therefore we defer this diff --git a/debian/patches/features/all/rt/slub-disable-SLUB_CPU_PARTIAL.patch b/debian/patches/features/all/rt/slub-disable-SLUB_CPU_PARTIAL.patch index 34f87bda8e32..c7c1bf6bbbcf 100644 --- a/debian/patches/features/all/rt/slub-disable-SLUB_CPU_PARTIAL.patch +++ b/debian/patches/features/all/rt/slub-disable-SLUB_CPU_PARTIAL.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Wed, 15 Apr 2015 19:00:47 +0200 Subject: slub: Disable SLUB_CPU_PARTIAL -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz |BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:915 |in_atomic(): 1, irqs_disabled(): 0, pid: 87, name: rcuop/7 diff --git a/debian/patches/features/all/rt/slub-enable-irqs-for-no-wait.patch b/debian/patches/features/all/rt/slub-enable-irqs-for-no-wait.patch index fde1878152f1..8d1c0a3baa11 100644 --- a/debian/patches/features/all/rt/slub-enable-irqs-for-no-wait.patch +++ b/debian/patches/features/all/rt/slub-enable-irqs-for-no-wait.patch @@ -1,7 +1,7 @@ Subject: slub: Enable irqs for __GFP_WAIT From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 09 Jan 2013 12:08:15 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz SYSTEM_RUNNING might be too late for enabling interrupts. Allocations with GFP_WAIT can happen before that. So use this as an indicator. diff --git a/debian/patches/features/all/rt/snd-pcm-fix-snd_pcm_stream_lock-irqs_disabled-splats.patch b/debian/patches/features/all/rt/snd-pcm-fix-snd_pcm_stream_lock-irqs_disabled-splats.patch index 94e79dcf8348..2de192b093f7 100644 --- a/debian/patches/features/all/rt/snd-pcm-fix-snd_pcm_stream_lock-irqs_disabled-splats.patch +++ b/debian/patches/features/all/rt/snd-pcm-fix-snd_pcm_stream_lock-irqs_disabled-splats.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com> Date: Wed, 18 Feb 2015 15:09:23 +0100 Subject: snd/pcm: fix snd_pcm_stream_lock*() irqs_disabled() splats -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Locking functions previously using read_lock_irq()/read_lock_irqsave() were changed to local_irq_disable/save(), leading to gripes. Use nort variants. diff --git a/debian/patches/features/all/rt/softirq-disable-softirq-stacks-for-rt.patch b/debian/patches/features/all/rt/softirq-disable-softirq-stacks-for-rt.patch index 16c0c7b1c440..7104d327760e 100644 --- a/debian/patches/features/all/rt/softirq-disable-softirq-stacks-for-rt.patch +++ b/debian/patches/features/all/rt/softirq-disable-softirq-stacks-for-rt.patch @@ -1,7 +1,7 @@ Subject: softirq: Disable softirq stacks for RT From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 18 Jul 2011 13:59:17 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Disable extra stacks for softirqs. We want to preempt softirqs and having them on special IRQ-stack does not make this easier. diff --git a/debian/patches/features/all/rt/softirq-preempt-fix-3-re.patch b/debian/patches/features/all/rt/softirq-preempt-fix-3-re.patch index 731987e49d34..5e4f011e1d65 100644 --- a/debian/patches/features/all/rt/softirq-preempt-fix-3-re.patch +++ b/debian/patches/features/all/rt/softirq-preempt-fix-3-re.patch @@ -1,7 +1,7 @@ Subject: softirq: Check preemption after reenabling interrupts From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 13 Nov 2011 17:17:09 +0100 (CET) -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz raise_softirq_irqoff() disables interrupts and wakes the softirq daemon, but after reenabling interrupts there is no preemption check, diff --git a/debian/patches/features/all/rt/softirq-split-locks.patch b/debian/patches/features/all/rt/softirq-split-locks.patch index fdbe4c907cad..3a5b1804d0a5 100644 --- a/debian/patches/features/all/rt/softirq-split-locks.patch +++ b/debian/patches/features/all/rt/softirq-split-locks.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 04 Oct 2012 14:20:47 +0100 Subject: softirq: Split softirq locks -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The 3.x RT series removed the split softirq implementation in favour of pushing softirq processing into the context of the thread which @@ -173,7 +173,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> * Are we in NMI context? --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -1832,6 +1832,8 @@ struct task_struct { +@@ -1833,6 +1833,8 @@ struct task_struct { #endif #ifdef CONFIG_PREEMPT_RT_BASE struct rcu_head put_rcu; @@ -182,7 +182,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> #endif #ifdef CONFIG_DEBUG_ATOMIC_SLEEP unsigned long task_state_change; -@@ -2096,6 +2098,7 @@ extern void thread_group_cputime_adjuste +@@ -2098,6 +2100,7 @@ extern void thread_group_cputime_adjuste /* * Per process flags */ diff --git a/debian/patches/features/all/rt/sparc64-use-generic-rwsem-spinlocks-rt.patch b/debian/patches/features/all/rt/sparc64-use-generic-rwsem-spinlocks-rt.patch index a1ff007c1baa..1f210646639e 100644 --- a/debian/patches/features/all/rt/sparc64-use-generic-rwsem-spinlocks-rt.patch +++ b/debian/patches/features/all/rt/sparc64-use-generic-rwsem-spinlocks-rt.patch @@ -2,7 +2,7 @@ From d6a6675d436897cd1b09e299436df3499abd753e Mon Sep 17 00:00:00 2001 From: Allen Pais <allen.pais@oracle.com> Date: Fri, 13 Dec 2013 09:44:41 +0530 Subject: [PATCH 1/3] sparc64: use generic rwsem spinlocks rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Signed-off-by: Allen Pais <allen.pais@oracle.com> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> diff --git a/debian/patches/features/all/rt/spinlock-types-separate-raw.patch b/debian/patches/features/all/rt/spinlock-types-separate-raw.patch index bbee2ba0a661..de4ffa9e7a90 100644 --- a/debian/patches/features/all/rt/spinlock-types-separate-raw.patch +++ b/debian/patches/features/all/rt/spinlock-types-separate-raw.patch @@ -1,7 +1,7 @@ Subject: spinlock: Split the lock types header From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 29 Jun 2011 19:34:01 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Split raw_spinlock into its own file and the remaining spinlock_t into its own non-RT header. The non-RT header will be replaced later by sleeping diff --git a/debian/patches/features/all/rt/stomp-machine-create-lg_global_trylock_relax-primiti.patch b/debian/patches/features/all/rt/stomp-machine-create-lg_global_trylock_relax-primiti.patch index d541192979e8..f20c39596e0c 100644 --- a/debian/patches/features/all/rt/stomp-machine-create-lg_global_trylock_relax-primiti.patch +++ b/debian/patches/features/all/rt/stomp-machine-create-lg_global_trylock_relax-primiti.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com> Date: Fri, 2 May 2014 13:13:22 +0200 Subject: stomp-machine: create lg_global_trylock_relax() primitive -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Create lg_global_trylock_relax() for use by stopper thread when it cannot schedule, to deal with stop_cpus_lock, which is now an lglock. diff --git a/debian/patches/features/all/rt/stomp-machine-use-lg_global_trylock_relax-to-dead-wi.patch b/debian/patches/features/all/rt/stomp-machine-use-lg_global_trylock_relax-to-dead-wi.patch index 9826cdc86a02..16010e2a0e13 100644 --- a/debian/patches/features/all/rt/stomp-machine-use-lg_global_trylock_relax-to-dead-wi.patch +++ b/debian/patches/features/all/rt/stomp-machine-use-lg_global_trylock_relax-to-dead-wi.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com> Date: Fri, 2 May 2014 13:13:34 +0200 Subject: stomp-machine: use lg_global_trylock_relax() to dead with stop_cpus_lock lglock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz If the stop machinery is called from inactive CPU we cannot use lg_global_lock(), because some other stomp machine invocation might be diff --git a/debian/patches/features/all/rt/stop-machine-raw-lock.patch b/debian/patches/features/all/rt/stop-machine-raw-lock.patch index 4b63e01c6349..6b3893946fc5 100644 --- a/debian/patches/features/all/rt/stop-machine-raw-lock.patch +++ b/debian/patches/features/all/rt/stop-machine-raw-lock.patch @@ -1,7 +1,7 @@ Subject: stop_machine: Use raw spinlocks From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 29 Jun 2011 11:01:51 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use raw-locks in stomp_machine() to allow locking in irq-off regions. diff --git a/debian/patches/features/all/rt/stop_machine-convert-stop_machine_run-to-PREEMPT_RT.patch b/debian/patches/features/all/rt/stop_machine-convert-stop_machine_run-to-PREEMPT_RT.patch index 098fa2f37945..90ca811a6ab3 100644 --- a/debian/patches/features/all/rt/stop_machine-convert-stop_machine_run-to-PREEMPT_RT.patch +++ b/debian/patches/features/all/rt/stop_machine-convert-stop_machine_run-to-PREEMPT_RT.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:30:27 -0500 Subject: stop_machine: convert stop_machine_run() to PREEMPT_RT -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Instead of playing with non-preemption, introduce explicit startup serialization. This is more robust and cleaner as diff --git a/debian/patches/features/all/rt/sunrpc-make-svc_xprt_do_enqueue-use-get_cpu_light.patch b/debian/patches/features/all/rt/sunrpc-make-svc_xprt_do_enqueue-use-get_cpu_light.patch index 4fe10aceafd3..8e4ce77210f6 100644 --- a/debian/patches/features/all/rt/sunrpc-make-svc_xprt_do_enqueue-use-get_cpu_light.patch +++ b/debian/patches/features/all/rt/sunrpc-make-svc_xprt_do_enqueue-use-get_cpu_light.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com> Date: Wed, 18 Feb 2015 16:05:28 +0100 Subject: sunrpc: Make svc_xprt_do_enqueue() use get_cpu_light() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz |BUG: sleeping function called from invalid context at kernel/locking/rtmutex.c:915 |in_atomic(): 1, irqs_disabled(): 0, pid: 3194, name: rpc.nfsd diff --git a/debian/patches/features/all/rt/suspend-prevernt-might-sleep-splats.patch b/debian/patches/features/all/rt/suspend-prevernt-might-sleep-splats.patch index 72b1d3da6678..502aaabc2cca 100644 --- a/debian/patches/features/all/rt/suspend-prevernt-might-sleep-splats.patch +++ b/debian/patches/features/all/rt/suspend-prevernt-might-sleep-splats.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 15 Jul 2010 10:29:00 +0200 Subject: suspend: Prevent might sleep splats -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz timekeeping suspend/resume calls read_persistant_clock() which takes rtc_lock. That results in might sleep warnings because at that point diff --git a/debian/patches/features/all/rt/sysfs-realtime-entry.patch b/debian/patches/features/all/rt/sysfs-realtime-entry.patch index 596ab735d095..4f08225534cf 100644 --- a/debian/patches/features/all/rt/sysfs-realtime-entry.patch +++ b/debian/patches/features/all/rt/sysfs-realtime-entry.patch @@ -1,7 +1,7 @@ Subject: sysfs: Add /sys/kernel/realtime entry From: Clark Williams <williams@redhat.com> Date: Sat Jul 30 21:55:53 2011 -0500 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Add a /sys/kernel entry to indicate that the kernel is a realtime kernel. diff --git a/debian/patches/features/all/rt/tasklet-rt-prevent-tasklets-from-going-into-infinite-spin-in-rt.patch b/debian/patches/features/all/rt/tasklet-rt-prevent-tasklets-from-going-into-infinite-spin-in-rt.patch index 908fb51a8045..498224e01965 100644 --- a/debian/patches/features/all/rt/tasklet-rt-prevent-tasklets-from-going-into-infinite-spin-in-rt.patch +++ b/debian/patches/features/all/rt/tasklet-rt-prevent-tasklets-from-going-into-infinite-spin-in-rt.patch @@ -1,7 +1,7 @@ Subject: tasklet: Prevent tasklets from going into infinite spin in RT From: Ingo Molnar <mingo@elte.hu> Date: Tue Nov 29 20:18:22 2011 -0500 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz When CONFIG_PREEMPT_RT_FULL is enabled, tasklets run as threads, and spinlocks turn are mutexes. But this can cause issues with diff --git a/debian/patches/features/all/rt/tasklist-lock-fix-section-conflict.patch b/debian/patches/features/all/rt/tasklist-lock-fix-section-conflict.patch index 453e6330f1df..dbeaaf569dfe 100644 --- a/debian/patches/features/all/rt/tasklist-lock-fix-section-conflict.patch +++ b/debian/patches/features/all/rt/tasklist-lock-fix-section-conflict.patch @@ -1,7 +1,7 @@ Subject: rwlocks: Fix section mismatch From: John Kacur <jkacur@redhat.com> Date: Mon, 19 Sep 2011 11:09:27 +0200 (CEST) -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz This fixes the following build error for the preempt-rt kernel. diff --git a/debian/patches/features/all/rt/thermal-Defer-thermal-wakups-to-threads.patch b/debian/patches/features/all/rt/thermal-Defer-thermal-wakups-to-threads.patch index 4aef42297f5a..b295b821b4f1 100644 --- a/debian/patches/features/all/rt/thermal-Defer-thermal-wakups-to-threads.patch +++ b/debian/patches/features/all/rt/thermal-Defer-thermal-wakups-to-threads.patch @@ -1,7 +1,7 @@ From: Daniel Wagner <wagi@monom.org> Date: Tue, 17 Feb 2015 09:37:44 +0100 Subject: thermal: Defer thermal wakups to threads -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On RT the spin lock in pkg_temp_thermal_platfrom_thermal_notify will call schedule while we run in irq context. diff --git a/debian/patches/features/all/rt/timekeeping-split-jiffies-lock.patch b/debian/patches/features/all/rt/timekeeping-split-jiffies-lock.patch index 2ef9a970f119..f05e39fd1aba 100644 --- a/debian/patches/features/all/rt/timekeeping-split-jiffies-lock.patch +++ b/debian/patches/features/all/rt/timekeeping-split-jiffies-lock.patch @@ -1,7 +1,7 @@ Subject: timekeeping: Split jiffies seqlock From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 14 Feb 2013 22:36:59 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Replace jiffies_lock seqlock with a simple seqcounter and a rawlock so it can be taken in atomic context on RT. diff --git a/debian/patches/features/all/rt/timer-delay-waking-softirqs-from-the-jiffy-tick.patch b/debian/patches/features/all/rt/timer-delay-waking-softirqs-from-the-jiffy-tick.patch index 51404838cd4f..3ee5bee0013c 100644 --- a/debian/patches/features/all/rt/timer-delay-waking-softirqs-from-the-jiffy-tick.patch +++ b/debian/patches/features/all/rt/timer-delay-waking-softirqs-from-the-jiffy-tick.patch @@ -1,7 +1,7 @@ From: Peter Zijlstra <peterz@infradead.org> Date: Fri, 21 Aug 2009 11:56:45 +0200 Subject: timer: delay waking softirqs from the jiffy tick -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz People were complaining about broken balancing with the recent -rt series. diff --git a/debian/patches/features/all/rt/timer-fd-avoid-live-lock.patch b/debian/patches/features/all/rt/timer-fd-avoid-live-lock.patch index 91924348a0a4..51d9179dd656 100644 --- a/debian/patches/features/all/rt/timer-fd-avoid-live-lock.patch +++ b/debian/patches/features/all/rt/timer-fd-avoid-live-lock.patch @@ -1,7 +1,7 @@ Subject: timer-fd: Prevent live lock From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 25 Jan 2012 11:08:40 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz If hrtimer_try_to_cancel() requires a retry, then depending on the priority setting te retry loop might prevent timer callback completion diff --git a/debian/patches/features/all/rt/timers-avoid-the-base-null-otptimization-on-rt.patch b/debian/patches/features/all/rt/timers-avoid-the-base-null-otptimization-on-rt.patch index afcd060f45ed..784a94995385 100644 --- a/debian/patches/features/all/rt/timers-avoid-the-base-null-otptimization-on-rt.patch +++ b/debian/patches/features/all/rt/timers-avoid-the-base-null-otptimization-on-rt.patch @@ -1,7 +1,7 @@ Subject: timers: Avoid the switch timers base set to NULL trick on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 21 Jul 2011 15:23:39 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz On RT that code is preemptible, so we cannot assign NULL to timers base as a preempter would spin forever in lock_timer_base(). diff --git a/debian/patches/features/all/rt/timers-preempt-rt-support.patch b/debian/patches/features/all/rt/timers-preempt-rt-support.patch index f4acca7021fa..4902adaa320b 100644 --- a/debian/patches/features/all/rt/timers-preempt-rt-support.patch +++ b/debian/patches/features/all/rt/timers-preempt-rt-support.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:30:20 -0500 Subject: timers: Preempt-rt support -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The base->lock is a sleeping lock. Try not to workaround with a spin_trylock(). The rt-mutex lock is not irq save even the try-lock diff --git a/debian/patches/features/all/rt/timers-prepare-for-full-preemption.patch b/debian/patches/features/all/rt/timers-prepare-for-full-preemption.patch index cae0f96bb642..bb41c11cee9c 100644 --- a/debian/patches/features/all/rt/timers-prepare-for-full-preemption.patch +++ b/debian/patches/features/all/rt/timers-prepare-for-full-preemption.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:34 -0500 Subject: timers: Prepare for full preemption -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz When softirqs can be preempted we need to make sure that cancelling the timer from the active thread can not deadlock vs. a running timer diff --git a/debian/patches/features/all/rt/tracing-account-for-preempt-off-in-preempt_schedule.patch b/debian/patches/features/all/rt/tracing-account-for-preempt-off-in-preempt_schedule.patch index 15f7c444c4c0..6283a8aa83bc 100644 --- a/debian/patches/features/all/rt/tracing-account-for-preempt-off-in-preempt_schedule.patch +++ b/debian/patches/features/all/rt/tracing-account-for-preempt-off-in-preempt_schedule.patch @@ -1,7 +1,7 @@ From: Steven Rostedt <rostedt@goodmis.org> Date: Thu, 29 Sep 2011 12:24:30 -0500 Subject: tracing: Account for preempt off in preempt_schedule() -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz The preempt_schedule() uses the preempt_disable_notrace() version because it can cause infinite recursion by the function tracer as diff --git a/debian/patches/features/all/rt/upstream-net-rt-remove-preemption-disabling-in-netif_rx.patch b/debian/patches/features/all/rt/upstream-net-rt-remove-preemption-disabling-in-netif_rx.patch index 7b69733dfb59..025ddeab2a31 100644 --- a/debian/patches/features/all/rt/upstream-net-rt-remove-preemption-disabling-in-netif_rx.patch +++ b/debian/patches/features/all/rt/upstream-net-rt-remove-preemption-disabling-in-netif_rx.patch @@ -1,7 +1,7 @@ Subject: net: Remove preemption disabling in netif_rx() From: Priyanka Jain <Priyanka.Jain@freescale.com> Date: Thu, 17 May 2012 09:35:11 +0530 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz 1)enqueue_to_backlog() (called from netif_rx) should be bind to a particluar CPU. This can be achieved by diff --git a/debian/patches/features/all/rt/usb-use-_nort-in-giveback.patch b/debian/patches/features/all/rt/usb-use-_nort-in-giveback.patch index d12d06a72efc..0bb72af50176 100644 --- a/debian/patches/features/all/rt/usb-use-_nort-in-giveback.patch +++ b/debian/patches/features/all/rt/usb-use-_nort-in-giveback.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Fri, 8 Nov 2013 17:34:54 +0100 Subject: usb: Use _nort in giveback function -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Since commit 94dfd7ed ("USB: HCD: support giveback of URB in tasklet context") I see diff --git a/debian/patches/features/all/rt/user-use-local-irq-nort.patch b/debian/patches/features/all/rt/user-use-local-irq-nort.patch index 65f511a27a27..fcf0a910a608 100644 --- a/debian/patches/features/all/rt/user-use-local-irq-nort.patch +++ b/debian/patches/features/all/rt/user-use-local-irq-nort.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 21 Jul 2009 23:06:05 +0200 Subject: core: Do not disable interrupts on RT in kernel/users.c -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use the local_irq_*_nort variants to reduce latencies in RT. The code is serialized by the locks. No need to disable interrupts. diff --git a/debian/patches/features/all/rt/vtime-split-lock-and-seqcount.patch b/debian/patches/features/all/rt/vtime-split-lock-and-seqcount.patch index fe4046b6d01a..e5d5723a7dca 100644 --- a/debian/patches/features/all/rt/vtime-split-lock-and-seqcount.patch +++ b/debian/patches/features/all/rt/vtime-split-lock-and-seqcount.patch @@ -1,7 +1,7 @@ Subject: vtime: Split lock and seqcount From: Thomas Gleixner <tglx@linutronix.de> Date: Tue, 23 Jul 2013 15:45:51 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Replace vtime_seqlock seqlock with a simple seqcounter and a rawlock so it can taken in atomic context on RT. @@ -27,7 +27,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> #else --- a/include/linux/sched.h +++ b/include/linux/sched.h -@@ -1519,7 +1519,8 @@ struct task_struct { +@@ -1520,7 +1520,8 @@ struct task_struct { cputime_t gtime; struct prev_cputime prev_cputime; #ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN diff --git a/debian/patches/features/all/rt/wait-simple-implementation.patch b/debian/patches/features/all/rt/wait-simple-implementation.patch index 5a0284cdaff5..3806f8f627e0 100644 --- a/debian/patches/features/all/rt/wait-simple-implementation.patch +++ b/debian/patches/features/all/rt/wait-simple-implementation.patch @@ -12,7 +12,7 @@ should go mainline as well, as it lowers memory consumption and runtime overhead. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz smp_mb() added by Steven Rostedt to fix a race condition with swait wakeups vs adding items to the list. diff --git a/debian/patches/features/all/rt/wait.h-include-atomic.h.patch b/debian/patches/features/all/rt/wait.h-include-atomic.h.patch index 8b04097cd1f1..bdfbb528d36e 100644 --- a/debian/patches/features/all/rt/wait.h-include-atomic.h.patch +++ b/debian/patches/features/all/rt/wait.h-include-atomic.h.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Mon, 28 Oct 2013 12:19:57 +0100 Subject: wait.h: include atomic.h -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz | CC init/main.o |In file included from include/linux/mmzone.h:9:0, diff --git a/debian/patches/features/all/rt/work-queue-work-around-irqsafe-timer-optimization.patch b/debian/patches/features/all/rt/work-queue-work-around-irqsafe-timer-optimization.patch index 3076491e95d1..f2fa2308c287 100644 --- a/debian/patches/features/all/rt/work-queue-work-around-irqsafe-timer-optimization.patch +++ b/debian/patches/features/all/rt/work-queue-work-around-irqsafe-timer-optimization.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 01 Jul 2013 11:02:42 +0200 Subject: workqueue: Prevent workqueue versus ata-piix livelock -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz An Intel i7 system regularly detected rcu_preempt stalls after the kernel was upgraded from 3.6-rt to 3.8-rt. When the stall happened, disk I/O was no diff --git a/debian/patches/features/all/rt/work-simple-Simple-work-queue-implemenation.patch b/debian/patches/features/all/rt/work-simple-Simple-work-queue-implemenation.patch index 447e301b0e2c..41f64e821186 100644 --- a/debian/patches/features/all/rt/work-simple-Simple-work-queue-implemenation.patch +++ b/debian/patches/features/all/rt/work-simple-Simple-work-queue-implemenation.patch @@ -1,7 +1,7 @@ From: Daniel Wagner <daniel.wagner@bmw-carit.de> Date: Fri, 11 Jul 2014 15:26:11 +0200 Subject: work-simple: Simple work queue implemenation -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Provides a framework for enqueuing callbacks from irq context PREEMPT_RT_FULL safe. The callbacks are executed in kthread context. diff --git a/debian/patches/features/all/rt/workqueue-distangle-from-rq-lock.patch b/debian/patches/features/all/rt/workqueue-distangle-from-rq-lock.patch index f6eff458e34e..05d0bee8bd53 100644 --- a/debian/patches/features/all/rt/workqueue-distangle-from-rq-lock.patch +++ b/debian/patches/features/all/rt/workqueue-distangle-from-rq-lock.patch @@ -22,7 +22,7 @@ Cc: Jens Axboe <axboe@kernel.dk> Cc: Linus Torvalds <torvalds@linux-foundation.org> Link: http://lkml.kernel.org/r/20110622174919.135236139@linutronix.de Signed-off-by: Thomas Gleixner <tglx@linutronix.de> -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz --- kernel/sched/core.c | 80 ++++++++------------------------------------ diff --git a/debian/patches/features/all/rt/workqueue-prevent-deadlock-stall.patch b/debian/patches/features/all/rt/workqueue-prevent-deadlock-stall.patch index b797367565dc..ccfea9697ad0 100644 --- a/debian/patches/features/all/rt/workqueue-prevent-deadlock-stall.patch +++ b/debian/patches/features/all/rt/workqueue-prevent-deadlock-stall.patch @@ -1,7 +1,7 @@ Subject: workqueue: Prevent deadlock/stall on RT From: Thomas Gleixner <tglx@linutronix.de> Date: Fri, 27 Jun 2014 16:24:52 +0200 (CEST) -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Austin reported a XFS deadlock/stall on RT where scheduled work gets never exececuted and tasks are waiting for each other for ever. diff --git a/debian/patches/features/all/rt/workqueue-use-locallock.patch b/debian/patches/features/all/rt/workqueue-use-locallock.patch index 3948dfed36b4..a062e30137c1 100644 --- a/debian/patches/features/all/rt/workqueue-use-locallock.patch +++ b/debian/patches/features/all/rt/workqueue-use-locallock.patch @@ -1,7 +1,7 @@ Subject: workqueue: Use local irq lock instead of irq disable regions From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 17 Jul 2011 21:42:26 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Use a local_irq_lock as a replacement for irq off regions. We keep the semantic of irq-off in regard to the pool->lock and remain preemptible. diff --git a/debian/patches/features/all/rt/workqueue-use-rcu.patch b/debian/patches/features/all/rt/workqueue-use-rcu.patch index 1511e9246767..9a78a0382a99 100644 --- a/debian/patches/features/all/rt/workqueue-use-rcu.patch +++ b/debian/patches/features/all/rt/workqueue-use-rcu.patch @@ -1,7 +1,7 @@ Subject: workqueue: Use normal rcu From: Thomas Gleixner <tglx@linutronix.de> Date: Wed, 24 Jul 2013 15:26:54 +0200 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz There is no need for sched_rcu. The undocumented reason why sched_rcu is used is to avoid a few explicit rcu_read_lock()/unlock() pairs by diff --git a/debian/patches/features/all/rt/x86-UV-raw_spinlock-conversion.patch b/debian/patches/features/all/rt/x86-UV-raw_spinlock-conversion.patch index b97b4ce614d2..f209c3b826fb 100644 --- a/debian/patches/features/all/rt/x86-UV-raw_spinlock-conversion.patch +++ b/debian/patches/features/all/rt/x86-UV-raw_spinlock-conversion.patch @@ -1,7 +1,7 @@ From: Mike Galbraith <umgwanakikbuti@gmail.com> Date: Sun, 2 Nov 2014 08:31:37 +0100 Subject: x86: UV: raw_spinlock conversion -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Shrug. Lots of hobbyists have a beast in their basement, right? diff --git a/debian/patches/features/all/rt/x86-crypto-reduce-preempt-disabled-regions.patch b/debian/patches/features/all/rt/x86-crypto-reduce-preempt-disabled-regions.patch index 71d9160a3942..833cd8c862c0 100644 --- a/debian/patches/features/all/rt/x86-crypto-reduce-preempt-disabled-regions.patch +++ b/debian/patches/features/all/rt/x86-crypto-reduce-preempt-disabled-regions.patch @@ -1,7 +1,7 @@ Subject: x86: crypto: Reduce preempt disabled regions From: Peter Zijlstra <peterz@infradead.org> Date: Mon, 14 Nov 2011 18:19:27 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Restrict the preempt disabled regions to the actual floating point operations and enable preemption for the administrative actions. diff --git a/debian/patches/features/all/rt/x86-highmem-add-a-already-used-pte-check.patch b/debian/patches/features/all/rt/x86-highmem-add-a-already-used-pte-check.patch index 477259650c72..d7c19fb6f0b5 100644 --- a/debian/patches/features/all/rt/x86-highmem-add-a-already-used-pte-check.patch +++ b/debian/patches/features/all/rt/x86-highmem-add-a-already-used-pte-check.patch @@ -1,7 +1,7 @@ From: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Date: Mon, 11 Mar 2013 17:09:55 +0100 Subject: x86/highmem: Add a "already used pte" check -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz This is a copy from kmap_atomic_prot(). diff --git a/debian/patches/features/all/rt/x86-io-apic-migra-no-unmask.patch b/debian/patches/features/all/rt/x86-io-apic-migra-no-unmask.patch index b17c57908ad7..1c1ae2d598a4 100644 --- a/debian/patches/features/all/rt/x86-io-apic-migra-no-unmask.patch +++ b/debian/patches/features/all/rt/x86-io-apic-migra-no-unmask.patch @@ -1,7 +1,7 @@ From: Ingo Molnar <mingo@elte.hu> Date: Fri, 3 Jul 2009 08:29:27 -0500 Subject: x86/ioapic: Do not unmask io_apic when interrupt is in progress -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz With threaded interrupts we might see an interrupt in progress on migration. Do not unmask it when this is the case. diff --git a/debian/patches/features/all/rt/x86-kvm-require-const-tsc-for-rt.patch b/debian/patches/features/all/rt/x86-kvm-require-const-tsc-for-rt.patch index 6b389598fddd..7cf458affc2b 100644 --- a/debian/patches/features/all/rt/x86-kvm-require-const-tsc-for-rt.patch +++ b/debian/patches/features/all/rt/x86-kvm-require-const-tsc-for-rt.patch @@ -1,7 +1,7 @@ Subject: x86: kvm Require const tsc for RT From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 06 Nov 2011 12:26:18 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Non constant TSC is a nightmare on bare metal already, but with virtualization it becomes a complete disaster because the workarounds @@ -15,7 +15,7 @@ Signed-off-by: Thomas Gleixner <tglx@linutronix.de> --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c -@@ -5782,6 +5782,13 @@ int kvm_arch_init(void *opaque) +@@ -5787,6 +5787,13 @@ int kvm_arch_init(void *opaque) goto out; } diff --git a/debian/patches/features/all/rt/x86-mce-timer-hrtimer.patch b/debian/patches/features/all/rt/x86-mce-timer-hrtimer.patch index bb31aa7ca181..612427b03bc6 100644 --- a/debian/patches/features/all/rt/x86-mce-timer-hrtimer.patch +++ b/debian/patches/features/all/rt/x86-mce-timer-hrtimer.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Mon, 13 Dec 2010 16:33:39 +0100 Subject: x86: Convert mce timer to hrtimer -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz mce_timer is started in atomic contexts of cpu bringup. This results in might_sleep() warnings on RT. Convert mce_timer to a hrtimer to @@ -35,7 +35,7 @@ fold in: #include <asm/processor.h> #include <asm/traps.h> -@@ -1225,7 +1226,7 @@ void mce_log_therm_throt_event(__u64 sta +@@ -1236,7 +1237,7 @@ void mce_log_therm_throt_event(__u64 sta static unsigned long check_interval = INITIAL_CHECK_INTERVAL; static DEFINE_PER_CPU(unsigned long, mce_next_interval); /* in jiffies */ @@ -44,7 +44,7 @@ fold in: static unsigned long mce_adjust_timer_default(unsigned long interval) { -@@ -1234,32 +1235,18 @@ static unsigned long mce_adjust_timer_de +@@ -1245,32 +1246,18 @@ static unsigned long mce_adjust_timer_de static unsigned long (*mce_adjust_timer)(unsigned long interval) = mce_adjust_timer_default; @@ -83,7 +83,7 @@ fold in: iv = __this_cpu_read(mce_next_interval); if (mce_available(this_cpu_ptr(&cpu_info))) { -@@ -1282,7 +1269,7 @@ static void mce_timer_fn(unsigned long d +@@ -1293,7 +1280,7 @@ static void mce_timer_fn(unsigned long d done: __this_cpu_write(mce_next_interval, iv); @@ -92,7 +92,7 @@ fold in: } /* -@@ -1290,7 +1277,7 @@ static void mce_timer_fn(unsigned long d +@@ -1301,7 +1288,7 @@ static void mce_timer_fn(unsigned long d */ void mce_timer_kick(unsigned long interval) { @@ -101,7 +101,7 @@ fold in: unsigned long iv = __this_cpu_read(mce_next_interval); __restart_timer(t, interval); -@@ -1305,7 +1292,7 @@ static void mce_timer_delete_all(void) +@@ -1316,7 +1303,7 @@ static void mce_timer_delete_all(void) int cpu; for_each_online_cpu(cpu) @@ -110,7 +110,7 @@ fold in: } static void mce_do_trigger(struct work_struct *work) -@@ -1628,7 +1615,7 @@ static void __mcheck_cpu_clear_vendor(st +@@ -1639,7 +1626,7 @@ static void __mcheck_cpu_clear_vendor(st } } @@ -119,7 +119,7 @@ fold in: { unsigned long iv = check_interval * HZ; -@@ -1637,16 +1624,17 @@ static void mce_start_timer(unsigned int +@@ -1648,16 +1635,17 @@ static void mce_start_timer(unsigned int per_cpu(mce_next_interval, cpu) = iv; @@ -141,7 +141,7 @@ fold in: mce_start_timer(cpu, t); } -@@ -2365,6 +2353,8 @@ static void mce_disable_cpu(void *h) +@@ -2376,6 +2364,8 @@ static void mce_disable_cpu(void *h) if (!mce_available(raw_cpu_ptr(&cpu_info))) return; @@ -150,7 +150,7 @@ fold in: if (!(action & CPU_TASKS_FROZEN)) cmci_clear(); -@@ -2387,6 +2377,7 @@ static void mce_reenable_cpu(void *h) +@@ -2398,6 +2388,7 @@ static void mce_reenable_cpu(void *h) if (b->init) wrmsrl(MSR_IA32_MCx_CTL(i), b->ctl); } @@ -158,7 +158,7 @@ fold in: } /* Get notified when a cpu comes on/off. Be hotplug friendly. */ -@@ -2394,7 +2385,6 @@ static int +@@ -2405,7 +2396,6 @@ static int mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) { unsigned int cpu = (unsigned long)hcpu; @@ -166,7 +166,7 @@ fold in: switch (action & ~CPU_TASKS_FROZEN) { case CPU_ONLINE: -@@ -2414,11 +2404,9 @@ mce_cpu_callback(struct notifier_block * +@@ -2425,11 +2415,9 @@ mce_cpu_callback(struct notifier_block * break; case CPU_DOWN_PREPARE: smp_call_function_single(cpu, mce_disable_cpu, &action, 1); diff --git a/debian/patches/features/all/rt/x86-mce-use-swait-queue-for-mce-wakeups.patch b/debian/patches/features/all/rt/x86-mce-use-swait-queue-for-mce-wakeups.patch index ff801ac28b14..d11b3f209dfe 100644 --- a/debian/patches/features/all/rt/x86-mce-use-swait-queue-for-mce-wakeups.patch +++ b/debian/patches/features/all/rt/x86-mce-use-swait-queue-for-mce-wakeups.patch @@ -1,7 +1,7 @@ Subject: x86/mce: use swait queue for mce wakeups From: Steven Rostedt <rostedt@goodmis.org> Date: Fri, 27 Feb 2015 15:20:37 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz We had a customer report a lockup on a 3.0-rt kernel that had the following backtrace: @@ -69,7 +69,7 @@ Signed-off-by: Daniel Wagner <daniel.wagner@bmw-carit.de> #include <asm/processor.h> #include <asm/traps.h> -@@ -1302,6 +1303,56 @@ static void mce_do_trigger(struct work_s +@@ -1313,6 +1314,56 @@ static void mce_do_trigger(struct work_s static DECLARE_WORK(mce_trigger_work, mce_do_trigger); @@ -126,7 +126,7 @@ Signed-off-by: Daniel Wagner <daniel.wagner@bmw-carit.de> /* * Notify the user(s) about new machine check events. * Can be called from interrupt context, but not from machine check/NMI -@@ -1309,19 +1360,8 @@ static DECLARE_WORK(mce_trigger_work, mc +@@ -1320,19 +1371,8 @@ static DECLARE_WORK(mce_trigger_work, mc */ int mce_notify_irq(void) { @@ -147,7 +147,7 @@ Signed-off-by: Daniel Wagner <daniel.wagner@bmw-carit.de> return 1; } return 0; -@@ -2445,6 +2485,10 @@ static __init int mcheck_init_device(voi +@@ -2456,6 +2496,10 @@ static __init int mcheck_init_device(voi goto err_out; } diff --git a/debian/patches/features/all/rt/x86-preempt-lazy.patch b/debian/patches/features/all/rt/x86-preempt-lazy.patch index 3001e025399b..a6422557bfbe 100644 --- a/debian/patches/features/all/rt/x86-preempt-lazy.patch +++ b/debian/patches/features/all/rt/x86-preempt-lazy.patch @@ -1,7 +1,7 @@ Subject: x86: Support for lazy preemption From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 01 Nov 2012 11:03:47 +0100 -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Implement the x86 pieces for lazy preempt. diff --git a/debian/patches/features/all/rt/x86-signal-delay-calling-signals-on-32bit.patch b/debian/patches/features/all/rt/x86-signal-delay-calling-signals-on-32bit.patch index d738d6db9948..f6062baf82dc 100644 --- a/debian/patches/features/all/rt/x86-signal-delay-calling-signals-on-32bit.patch +++ b/debian/patches/features/all/rt/x86-signal-delay-calling-signals-on-32bit.patch @@ -1,7 +1,7 @@ From: Yang Shi <yang.shi@linaro.org> Date: Thu, 10 Dec 2015 10:58:51 -0800 Subject: x86/signal: delay calling signals on 32bit -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz When running some ptrace single step tests on x86-32 machine, the below problem is triggered: diff --git a/debian/patches/features/all/rt/x86-stackprot-no-random-on-rt.patch b/debian/patches/features/all/rt/x86-stackprot-no-random-on-rt.patch index 1e26769f543a..43a60e49570a 100644 --- a/debian/patches/features/all/rt/x86-stackprot-no-random-on-rt.patch +++ b/debian/patches/features/all/rt/x86-stackprot-no-random-on-rt.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Thu, 16 Dec 2010 14:25:18 +0100 Subject: x86: stackprotector: Avoid random pool on rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz CPU bringup calls into the random pool to initialize the stack canary. During boot that works nicely even on RT as the might sleep diff --git a/debian/patches/features/all/rt/x86-use-gen-rwsem-spinlocks-rt.patch b/debian/patches/features/all/rt/x86-use-gen-rwsem-spinlocks-rt.patch index 5701b68d12a0..8bf1b02f15ca 100644 --- a/debian/patches/features/all/rt/x86-use-gen-rwsem-spinlocks-rt.patch +++ b/debian/patches/features/all/rt/x86-use-gen-rwsem-spinlocks-rt.patch @@ -1,7 +1,7 @@ From: Thomas Gleixner <tglx@linutronix.de> Date: Sun, 26 Jul 2009 02:21:32 +0200 Subject: x86: Use generic rwsem_spinlocks on -rt -Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4-rc6-rt1/patches-4.4-rc6-rt1.tar.xz +Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.4/patches-4.4-rt2.tar.xz Simplifies the separation of anon_rw_semaphores and rw_semaphores for -rt. diff --git a/debian/patches/series-rt b/debian/patches/series-rt index 6c26cd38b676..b72e03e8dd38 100644 --- a/debian/patches/series-rt +++ b/debian/patches/series-rt @@ -13,7 +13,6 @@ ############################################################ # Stuff broken upstream, patches submitted ############################################################ -features/all/rt/sched-reset-task-s-lockless-wake-queues-on-fork.patch ############################################################ # Stuff which needs addressing upstream, but requires more |