user/sven/linux.git/kernel/softirq.c, branch v4.19.312

nohz: Fix missing tick reprogram when interrupting an inline softirq

2018-08-03T13:52:10Z

The full nohz tick is reprogrammed in irq_exit() only if the exit is not in a nesting interrupt. This stands as an optimization: whether a hardirq or a softirq is interrupted, the tick is going to be reprogrammed when necessary at the end of the inner interrupt, with even potential new updates on the timer queue. When soft interrupts are interrupted, it's assumed that they are executing on the tail of an interrupt return. In that case tick_nohz_irq_exit() is called after softirq processing to take care of the tick reprogramming. But the assumption is wrong: softirqs can be processed inline as well, ie: outside of an interrupt, like in a call to local_bh_enable() or from ksoftirqd. Inline softirqs don't reprogram the tick once they are done, as opposed to interrupt tail softirq processing. So if a tick interrupts an inline softirq processing, the next timer will neither be reprogrammed from the interrupting tick's irq_exit() nor after the interrupted softirq processing. This situation may leave the tick unprogrammed while timers are armed. To fix this, simply keep reprogramming the tick even if a softirq has been interrupted. That can be optimized further, but for now correctness is more important. Note that new timers enqueued in nohz_full mode after a softirq gets interrupted will still be handled just fine through self-IPIs triggered by the timer code. Reported-by: Anna-Maria Gleixner Signed-off-by: Frederic Weisbecker Signed-off-by: Thomas Gleixner Tested-by: Anna-Maria Gleixner Cc: stable@vger.kernel.org # 4.14+ Link: https://lkml.kernel.org/r/1533303094-15855-1-git-send-email-frederic@kernel.org

Mark HI and TASKLET softirq synchronous

2018-07-17T18:12:43Z

Way back in 4.9, we committed 4cd13c21b207 ("softirq: Let ksoftirqd do its job"), and ever since we've had small nagging issues with it. For example, we've had: 1ff688209e2e ("watchdog: core: make sure the watchdog_worker is not deferred") 8d5755b3f77b ("watchdog: softdog: fire watchdog even if softirqs do not get to run") 217f69743681 ("net: busy-poll: allow preemption in sk_busy_loop()") all of which worked around some of the effects of that commit. The DVB people have also complained that the commit causes excessive USB URB latencies, which seems to be due to the USB code using tasklets to schedule USB traffic. This seems to be an issue mainly when already living on the edge, but waiting for ksoftirqd to handle it really does seem to cause excessive latencies. Now Hanna Hawa reports that this issue isn't just limited to USB URB and DVB, but also causes timeout problems for the Marvell SoC team: "I'm facing kernel panic issue while running raid 5 on sata disks connected to Macchiatobin (Marvell community board with Armada-8040 SoC with 4 ARMv8 cores of CA72) Raid 5 built with Marvell DMA engine and async_tx mechanism (ASYNC_TX_DMA [=y]); the DMA driver (mv_xor_v2) uses a tasklet to clean the done descriptors from the queue" The latency problem causes a panic: mv_xor_v2 f0400000.xor: dma_sync_wait: timeout! Kernel panic - not syncing: async_tx_quiesce: DMA error waiting for transaction We've discussed simply just reverting the original commit entirely, and also much more involved solutions (with per-softirq threads etc). This patch is intentionally stupid and fairly limited, because the issue still remains, and the other solutions either got sidetracked or had other issues. We should probably also consider the timer softirqs to be synchronous and not be delayed to ksoftirqd (since they were the issue with the earlier watchdog problems), but that should be done as a separate patch. This does only the tasklet cases. Reported-and-tested-by: Hanna Hawa Reported-and-tested-by: Josef Griebichler Reported-by: Mauro Carvalho Chehab Cc: Alan Stern Cc: Greg Kroah-Hartman Cc: Eric Dumazet Cc: Ingo Molnar Signed-off-by: Linus Torvalds

softirq: Reorder trace_softirqs_on to prevent lockdep splat

2018-06-21T19:12:43Z

I'm able to reproduce a lockdep splat with config options: CONFIG_PROVE_LOCKING=y, CONFIG_DEBUG_LOCK_ALLOC=y and CONFIG_PREEMPTIRQ_EVENTS=y $ echo 1 > /d/tracing/events/preemptirq/preempt_enable/enable [ 26.112609] DEBUG_LOCKS_WARN_ON(current->softirqs_enabled) [ 26.112636] WARNING: CPU: 0 PID: 118 at kernel/locking/lockdep.c:3854 [...] [ 26.144229] Call Trace: [ 26.144926] [ 26.145506] lock_acquire+0x55/0x1b0 [ 26.146499] ? __do_softirq+0x46f/0x4d9 [ 26.147571] ? __do_softirq+0x46f/0x4d9 [ 26.148646] trace_preempt_on+0x8f/0x240 [ 26.149744] ? trace_preempt_on+0x4d/0x240 [ 26.150862] ? __do_softirq+0x46f/0x4d9 [ 26.151930] preempt_count_sub+0x18a/0x1a0 [ 26.152985] __do_softirq+0x46f/0x4d9 [ 26.153937] irq_exit+0x68/0xe0 [ 26.154755] smp_apic_timer_interrupt+0x271/0x280 [ 26.156056] apic_timer_interrupt+0xf/0x20 [ 26.157105] The issue was this: preempt_count = 1 << SOFTIRQ_SHIFT __local_bh_enable(cnt = 1 << SOFTIRQ_SHIFT) { if (softirq_count() == (cnt && SOFTIRQ_MASK)) { trace_softirqs_on() { current->softirqs_enabled = 1; } } preempt_count_sub(cnt) { trace_preempt_on() { tracepoint() { rcu_read_lock_sched() { // jumps into lockdep Where preempt_count still has softirqs disabled, but current->softirqs_enabled is true, and we get a splat. Link: http://lkml.kernel.org/r/20180607201143.247775-1-joel@joelfernandes.org Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Linus Torvalds Cc: Mathieu Desnoyers Cc: Tom Zanussi Cc: Namhyung Kim Cc: Thomas Glexiner Cc: Boqun Feng Cc: Paul McKenney Cc: Masami Hiramatsu Cc: Todd Kjos Cc: Erick Reyes Cc: Julia Cartwright Cc: Byungchul Park Cc: stable@vger.kernel.org Reviewed-by: Steven Rostedt (VMware) Fixes: d59158162e032 ("tracing: Add support for preempt and irq enable/disable events") Signed-off-by: Joel Fernandes (Google) Signed-off-by: Steven Rostedt (VMware)

Merge branch 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2018-06-05T02:59:22Z

Pull irq updates from Thomas Gleixner: - Consolidation of softirq pending: The softirq mask and its accessors/mutators have many implementations scattered around many architectures. Most do the same things consisting in a field in a per-cpu struct (often irq_cpustat_t) accessed through per-cpu ops. We can provide instead a generic efficient version that most of them can use. In fact s390 is the only exception because the field is stored in lowcore. - Support for level!?! triggered MSI (ARM) Over the past couple of years, we've seen some SoCs coming up with ways of signalling level interrupts using a new flavor of MSIs, where the MSI controller uses two distinct messages: one that raises a virtual line, and one that lowers it. The target MSI controller is in charge of maintaining the state of the line. This allows for a much simplified HW signal routing (no need to have hundreds of discrete lines to signal level interrupts if you already have a memory bus), but results in a departure from the current idea the kernel has of MSIs. - Support for Meson-AXG GPIO irqchip - Large stm32 irqchip rework (suspend/resume, hierarchical domains) - More SPDX conversions * 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (36 commits) ARM: dts: stm32: Add exti support to stm32mp157 pinctrl ARM: dts: stm32: Add exti support for stm32mp157c pinctrl/stm32: Add irq_eoi for stm32gpio irqchip irqchip/stm32: Add suspend/resume support for hierarchy domain irqchip/stm32: Add stm32mp1 support with hierarchy domain irqchip/stm32: Prepare common functions irqchip/stm32: Add host and driver data structures irqchip/stm32: Add suspend support irqchip/stm32: Add falling pending register support irqchip/stm32: Checkpatch fix irqchip/stm32: Optimizes and cleans up stm32-exti irq_domain irqchip/meson-gpio: Add support for Meson-AXG SoCs dt-bindings: interrupt-controller: New binding for Meson-AXG SoC dt-bindings: interrupt-controller: Fix the double quotes softirq/s390: Move default mutators of overwritten softirq mask to s390 softirq/x86: Switch to generic local_softirq_pending() implementation softirq/sparc: Switch to generic local_softirq_pending() implementation softirq/powerpc: Switch to generic local_softirq_pending() implementation softirq/parisc: Switch to generic local_softirq_pending() implementation softirq/ia64: Switch to generic local_softirq_pending() implementation ...

softirq: Eliminate unused cond_resched_softirq() macro

2018-05-15T17:27:35Z

The cond_resched_softirq() macro is not used anywhere in mainline, so this commit simplifies the kernel by eliminating it. Suggested-by: Eric Dumazet Signed-off-by: Paul E. McKenney Cc: Ingo Molnar Acked-by: Peter Zijlstra (Intel) Reviewed-by: Eric Dumazet Tested-by: Nicholas Piggin

softirq/core: Turn default irq_cpustat_t to standard per-cpu

2018-05-14T09:25:27Z

In order to optimize and consolidate softirq mask accesses, let's convert the default irq_cpustat_t implementation to per-CPU standard API. Signed-off-by: Frederic Weisbecker Acked-by: Thomas Gleixner Acked-by: Peter Zijlstra Cc: Benjamin Herrenschmidt Cc: David S. Miller Cc: Fenghua Yu Cc: Heiko Carstens Cc: Helge Deller Cc: James E.J. Bottomley Cc: Linus Torvalds Cc: Martin Schwidefsky Cc: Michael Ellerman Cc: Paul Mackerras Cc: Rich Felker Cc: Sebastian Andrzej Siewior Cc: Tony Luck Cc: Yoshinori Sato Link: http://lkml.kernel.org/r/1525786706-22846-5-git-send-email-frederic@kernel.org Signed-off-by: Ingo Molnar

softirq: Consolidate common code in tasklet_[hi]_action()

2018-03-09T10:50:55Z

tasklet_action() + tasklet_hi_action() are almost identical. Move the common code from both function into __tasklet_action_common() and let both functions invoke it with different arguments. [ bigeasy: Splitted out from RT's "tasklet: Prevent tasklets from going into infinite spin in RT" and added commit message] Signed-off-by: Ingo Molnar Signed-off-by: Steven Rostedt Signed-off-by: Thomas Gleixner Signed-off-by: Sebastian Andrzej Siewior Cc: Julia Cartwright Link: https://lkml.kernel.org/r/20180227164808.10093-3-bigeasy@linutronix.de

softirq: Consolidate common code in __tasklet_[hi]_schedule()

2018-03-09T10:50:55Z

__tasklet_schedule() and __tasklet_hi_schedule() are almost identical. Move the common code from both function into __tasklet_schedule_common() and let both functions invoke it with different arguments. [ bigeasy: Splitted out from RT's "tasklet: Prevent tasklets from going into infinite spin in RT" and added commit message. Use this_cpu_ptr(headp) in __tasklet_schedule_common() as suggested by Julia Cartwright ] Signed-off-by: Ingo Molnar Signed-off-by: Steven Rostedt Signed-off-by: Thomas Gleixner Signed-off-by: Sebastian Andrzej Siewior Cc: Julia Cartwright Link: https://lkml.kernel.org/r/20180227164808.10093-2-bigeasy@linutronix.de

softirq: Eliminate cond_resched_rcu_qs() in favor of cond_resched()

2017-12-04T18:28:58Z

Now that cond_resched() also provides RCU quiescent states when needed, it can be used in place of cond_resched_rcu_qs(). This commit therefore makes this change. Signed-off-by: Paul E. McKenney Cc: NeilBrown Cc: Ingo Molnar

kmemcheck: rip it out

2017-11-16T02:21:05Z

Fix up makefiles, remove references, and git rm kmemcheck. Link: http://lkml.kernel.org/r/20171007030159.22241-4-alexander.levin@verizon.com Signed-off-by: Sasha Levin Cc: Steven Rostedt Cc: Vegard Nossum Cc: Pekka Enberg Cc: Michal Hocko Cc: Eric W. Biederman Cc: Alexander Potapenko Cc: Tim Hansen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds