kernel.git/kernel/rcu/tree.c, branch linux-5.19.y

rcu: Back off upon fill_page_cache_func() allocation failure

2022-10-24T07:58:08Z

[ Upstream commit 093590c16b447f53e66771c8579ae66c96f6ef61 ] The fill_page_cache_func() function allocates couple of pages to store kvfree_rcu_bulk_data structures. This is a lightweight (GFP_NORETRY) allocation which can fail under memory pressure. The function will, however keep retrying even when the previous attempt has failed. This retrying is in theory correct, but in practice the allocation is invoked from workqueue context, which means that if the memory reclaim gets stuck, these retries can hog the worker for quite some time. Although the workqueues subsystem automatically adjusts concurrency, such adjustment is not guaranteed to happen until the worker context sleeps. And the fill_page_cache_func() function's retry loop is not guaranteed to sleep (see the should_reclaim_retry() function). And we have seen this function cause workqueue lockups: kernel: BUG: workqueue lockup - pool cpus=93 node=1 flags=0x1 nice=0 stuck for 32s! [...] kernel: pool 74: cpus=37 node=0 flags=0x1 nice=0 hung=32s workers=2 manager: 2146 kernel: pwq 498: cpus=249 node=1 flags=0x1 nice=0 active=4/256 refcnt=5 kernel: in-flight: 1917:fill_page_cache_func kernel: pending: dbs_work_handler, free_work, kfree_rcu_monitor Originally, we thought that the root cause of this lockup was several retries with direct reclaim, but this is not yet confirmed. Furthermore, we have seen similar lockups without any heavy memory pressure. This suggests that there are other factors contributing to these lockups. However, it is not really clear that endless retries are desireable. So let's make the fill_page_cache_func() function back off after allocation failure. Cc: Uladzislau Rezki (Sony) Cc: "Paul E. McKenney" Cc: Frederic Weisbecker Cc: Neeraj Upadhyay Cc: Josh Triplett Cc: Steven Rostedt Cc: Mathieu Desnoyers Cc: Lai Jiangshan Cc: Joel Fernandes Signed-off-by: Michal Hocko Reviewed-by: Uladzislau Rezki (Sony) Signed-off-by: Paul E. McKenney Signed-off-by: Sasha Levin

Merge branch 'exp.2022.05.11a' into HEAD

2022-05-11T18:49:35Z

exp.2022.05.11a: Expedited-grace-period latency-reduction updates.

rcu: Move expedited grace period (GP) work to RT kthread_worker

2022-05-11T18:47:10Z

Enabling CONFIG_RCU_BOOST did not reduce RCU expedited grace-period latency because its workqueues run at SCHED_OTHER, and thus can be delayed by normal processes. This commit avoids these delays by moving the expedited GP work items to a real-time-priority kthread_worker. This option is controlled by CONFIG_RCU_EXP_KTHREAD and disabled by default on PREEMPT_RT=y kernels which disable expedited grace periods after boot by unconditionally setting rcupdate.rcu_normal_after_boot=1. The results were evaluated on arm64 Android devices (6GB ram) running 5.10 kernel, and capturing trace data in critical user-level code. The table below shows the resulting order-of-magnitude improvements in synchronize_rcu_expedited() latency: ------------------------------------------------------------------------ | | workqueues | kthread_worker | Diff | ------------------------------------------------------------------------ | Count | 725 | 688 | | ------------------------------------------------------------------------ | Min Duration (ns) | 326 | 447 | 37.12% | ------------------------------------------------------------------------ | Q1 (ns) | 39,428 | 38,971 | -1.16% | ------------------------------------------------------------------------ | Q2 - Median (ns) | 98,225 | 69,743 | -29.00% | ------------------------------------------------------------------------ | Q3 (ns) | 342,122 | 126,638 | -62.98% | ------------------------------------------------------------------------ | Max Duration (ns) | 372,766,967 | 2,329,671 | -99.38% | ------------------------------------------------------------------------ | Avg Duration (ns) | 2,746,353 | 151,242 | -94.49% | ------------------------------------------------------------------------ | Standard Deviation (ns) | 19,327,765 | 294,408 | | ------------------------------------------------------------------------ The below table show the range of maximums/minimums for synchronize_rcu_expedited() latency from all experiments: ------------------------------------------------------------------------ | | workqueues | kthread_worker | Diff | ------------------------------------------------------------------------ | Total No. of Experiments | 25 | 23 | | ------------------------------------------------------------------------ | Largest Maximum (ns) | 372,766,967 | 2,329,671 | -99.38% | ------------------------------------------------------------------------ | Smallest Maximum (ns) | 38,819 | 86,954 | 124.00% | ------------------------------------------------------------------------ | Range of Maximums (ns) | 372,728,148 | 2,242,717 | | ------------------------------------------------------------------------ | Largest Minimum (ns) | 88,623 | 27,588 | -68.87% | ------------------------------------------------------------------------ | Smallest Minimum (ns) | 326 | 447 | 37.12% | ------------------------------------------------------------------------ | Range of Minimums (ns) | 88,297 | 27,141 | | ------------------------------------------------------------------------ Cc: "Paul E. McKenney" Cc: Tejun Heo Reported-by: Tim Murray Reported-by: Wei Wang Tested-by: Kyle Lin Tested-by: Chunwei Lu Tested-by: Lulu Wang Signed-off-by: Kalesh Singh Signed-off-by: Paul E. McKenney

Merge branches 'docs.2022.04.20a', 'fixes.2022.04.20a', 'nocb.2022.04.11b', 'rcu-tasks.2022.04.11b', 'srcu.2022.05.03a', 'torture.2022.04.11b', 'torture-tasks.2022.04.20a' and 'torturescript.2022.04.20a' into HEAD

2022-05-03T17:21:40Z

docs.2022.04.20a: Documentation updates. fixes.2022.04.20a: Miscellaneous fixes. nocb.2022.04.11b: Callback-offloading updates. rcu-tasks.2022.04.11b: RCU-tasks updates. srcu.2022.05.03a: Put SRCU on a memory diet. torture.2022.04.11b: Torture-test updates. torture-tasks.2022.04.20a: Avoid torture testing changing RCU configuration. torturescript.2022.04.20a: Torture-test scripting updates.

rcu: Fix preemption mode check on synchronize_rcu[_expedited]()

2022-04-20T23:51:11Z

An early check on synchronize_rcu[_expedited]() tries to determine if the current CPU is in UP mode on an SMP no-preempt kernel, in which case there is no need to start a grace period since the current assumed quiescent state is all we need. However the preemption mode doesn't take into account the boot selected preemption mode under CONFIG_PREEMPT_DYNAMIC=y, missing a possible early return if the running flavour is "none" or "voluntary". Use the shiny new preempt mode accessors to fix this. However, avoid invoking them during early boot because doing so triggers a WARN_ON_ONCE(). [ paulmck: Update for mainlined API. ] Reported-by: Paul E. McKenney Signed-off-by: Frederic Weisbecker Cc: Uladzislau Rezki Cc: Joel Fernandes Cc: Boqun Feng Cc: Peter Zijlstra Cc: Neeraj Upadhyay Cc: Valentin Schneider Signed-off-by: Paul E. McKenney

rcu: Add comments to final rcu_gp_cleanup() "if" statement

2022-04-20T23:51:10Z

The final "if" statement in rcu_gp_cleanup() has proven to be rather confusing, straightforward though it might have seemed when initially written. This commit therefore adds comments to its "then" and "else" clauses to at least provide a more elevated form of confusion. Reported-by: Boqun Feng Reported-by: Frederic Weisbecker Reported-by: Neeraj Upadhyay Reported-by: Uladzislau Rezki Signed-off-by: Paul E. McKenney

rcu: Check for jiffies going backwards

2022-04-12T00:28:48Z

A report of a 12-jiffy normal RCU CPU stall warning raises interesting questions about the nature of time on the offending system. This commit instruments rcu_sched_clock_irq(), which is RCU's hook into the scheduling-clock interrupt, checking for the jiffies counter going backwards. Reported-by: Saravanan D Signed-off-by: Paul E. McKenney

rcutorture: Suppress debugging grace period delays during flooding

2022-04-12T00:07:28Z

Tree RCU supports grace-period delays using the rcutree.gp_cleanup_delay, rcutree.gp_init_delay, and rcutree.gp_preinit_delay kernel boot parameters. These delays are strictly for debugging purposes, and have proven quite effective at exposing bugs involving race with CPU-hotplug operations. However, these delays can result in false positives when used in conjunction with callback flooding, for example, those generated by the rcutorture.fwd_progress kernel boot parameter. This commit therefore suppresses grace-period delays while callback flooding is in progress. Signed-off-by: Paul E. McKenney

rcu-tasks: Make Tasks RCU account for userspace execution

2022-04-12T00:06:42Z

The main Tasks RCU quiescent state is voluntary context switch. However, userspace execution is also a valid quiescent state, and is a valuable one for userspace applications that spin repeatedly executing light-weight non-sleeping system calls. Currently, such an application can delay a Tasks RCU grace period for many tens of seconds. This commit therefore enlists the aid of the scheduler-clock interrupt to provide a Tasks RCU quiescent state when it interrupted a task executing in userspace. [ paulmck: Apply feedback from kernel test robot. ] Cc: Martin KaFai Lau Cc: Neil Spring Signed-off-by: Paul E. McKenney

rcu/nocb: Initialize nocb kthreads only for boot CPU prior SMP initialization

2022-04-12T00:05:58Z

The rcu_spawn_gp_kthread() function is called as an early initcall, which means that SMP initialization hasn't happened yet and only the boot CPU is online. Therefore, create only the NOCB kthreads related to the boot CPU. Signed-off-by: Frederic Weisbecker Cc: Neeraj Upadhyay Cc: Uladzislau Rezki Cc: Joel Fernandes Cc: Boqun Feng Signed-off-by: Paul E. McKenney