linux/kernel/rcu/srcutree.c, branch v6.0

srcu: Make expedited RCU grace periods block even less frequently

2022-07-19T18:39:59Z

The purpose of commit 282d8998e997 ("srcu: Prevent expedited GPs and blocking readers from consuming CPU") was to prevent a long series of never-blocking expedited SRCU grace periods from blocking kernel-live-patching (KLP) progress. Although it was successful, it also resulted in excessive boot times on certain embedded workloads running under qemu with the "-bios QEMU_EFI.fd" command line. Here "excessive" means increasing the boot time up into the three-to-four minute range. This increase in boot time was due to the more than 6000 back-to-back invocations of synchronize_rcu_expedited() within the KVM host OS, which in turn resulted from qemu's emulation of a long series of MMIO accesses. Commit 640a7d37c3f4 ("srcu: Block less aggressively for expedited grace periods") did not significantly help this particular use case. Zhangfei Gao and Shameerali Kolothum Thodi did experiments varying the value of SRCU_MAX_NODELAY_PHASE with HZ=250 and with various values of non-sleeping per phase counts on a system with preemption enabled, and observed the following boot times: +──────────────────────────+────────────────+ | SRCU_MAX_NODELAY_PHASE | Boot time (s) | +──────────────────────────+────────────────+ | 100 | 30.053 | | 150 | 25.151 | | 200 | 20.704 | | 250 | 15.748 | | 500 | 11.401 | | 1000 | 11.443 | | 10000 | 11.258 | | 1000000 | 11.154 | +──────────────────────────+────────────────+ Analysis on the experiment results show additional improvements with CPU-bound delays approaching one jiffy in duration. This improvement was also seen when number of per-phase iterations were scaled to one jiffy. This commit therefore scales per-grace-period phase number of non-sleeping polls so that non-sleeping polls extend for about one jiffy. In addition, the delay-calculation call to srcu_get_delay() in srcu_gp_end() is replaced with a simple check for an expedited grace period. This change schedules callback invocation immediately after expedited grace periods complete, which results in greatly improved boot times. Testing done by Marc and Zhangfei confirms that this change recovers most of the performance degradation in boottime; for CONFIG_HZ_250 configuration, specifically, boot times improve from 3m50s to 41s on Marc's setup; and from 2m40s to ~9.7s on Zhangfei's setup. In addition to the changes to default per phase delays, this change adds 3 new kernel parameters - srcutree.srcu_max_nodelay, srcutree.srcu_max_nodelay_phase, and srcutree.srcu_retry_check_delay. This allows users to configure the srcu grace period scanning delays in order to more quickly react to additional use cases. Fixes: 640a7d37c3f4 ("srcu: Block less aggressively for expedited grace periods") Fixes: 282d8998e997 ("srcu: Prevent expedited GPs and blocking readers from consuming CPU") Reported-by: Zhangfei Gao Reported-by: yueluck Signed-off-by: Neeraj Upadhyay Tested-by: Marc Zyngier Tested-by: Zhangfei Gao Link: https://lore.kernel.org/all/20615615-0013-5adc-584f-2b1d5c03ebfc@linaro.org/ Signed-off-by: Paul E. McKenney

srcu: Block less aggressively for expedited grace periods

2022-07-19T18:39:59Z

Commit 282d8998e997 ("srcu: Prevent expedited GPs and blocking readers from consuming CPU") fixed a problem where a long-running expedited SRCU grace period could block kernel live patching. It did so by giving up on expediting once a given SRCU expedited grace period grew too old. Unfortunately, this added excessive delays to boots of virtual embedded systems specifying "-bios QEMU_EFI.fd" to qemu. This commit therefore makes the transition away from expediting less aggressive, increasing the per-grace-period phase number of non-sleeping polls of readers from one to three and increasing the required grace-period age from one jiffy (actually from zero to one jiffies) to two jiffies (actually from one to two jiffies). Fixes: 282d8998e997 ("srcu: Prevent expedited GPs and blocking readers from consuming CPU") Signed-off-by: Paul E. McKenney Reported-by: Zhangfei Gao Reported-by: chenxiang (M)" Cc: Shameerali Kolothum Thodi Cc: Paolo Bonzini Reviewed-by: Neeraj Upadhyay Link: https://lore.kernel.org/all/20615615-0013-5adc-584f-2b1d5c03ebfc@linaro.org/

srcu: Drop needless initialization of sdp in srcu_gp_start()

2022-05-03T17:20:57Z

Commit 9c7ef4c30f12 ("srcu: Make Tree SRCU able to operate without snp_node array") initializes the local variable sdp differently depending on the srcu's state in srcu_gp_start(). Either way, this initialization overwrites the value used when sdp is defined. This commit therefore drops this pointless definition-time initialization. Although there is no functional change, compiler code generation may be affected. Signed-off-by: Lukas Bulwahn Signed-off-by: Paul E. McKenney

srcu: Prevent expedited GPs and blocking readers from consuming CPU

2022-05-03T17:20:57Z

If an SRCU reader blocks while a synchronize_srcu_expedited() waits for that same reader, then that grace period will spawn an endless series of workqueue handlers, consuming a full CPU. This quickly gets pointless because consuming more CPU isn't going to make that reader get done faster, especially if it is blocked waiting for an external event. This commit therefore spawns at most one pair of back-to-back workqueue handlers per expedited grace period phase, instead inserting increasing delays as that grace period phase grows older, but capped at 10 jiffies. In any case, if there have been at least 100 back-to-back workqueue handlers within a single jiffy, regardless of grace period or grace-period phase, then a one-jiffy delay is inserted. [ paulmck: Apply feedback from kernel test robot. ] Cc: Neeraj Upadhyay Reported-by: Song Liu Tested-by: kernel test robot Signed-off-by: Paul E. McKenney

srcu: Add contention check to call_srcu() srcu_data ->lock acquisition

2022-05-03T17:20:57Z

This commit increases the sensitivity of contention detection by adding checks to the acquisition of the srcu_data structure's lock on the call_srcu() code path. Co-developed-by: Neeraj Upadhyay Signed-off-by: Neeraj Upadhyay Signed-off-by: Paul E. McKenney

srcu: Automatically determine size-transition strategy at boot

2022-05-03T17:19:39Z

This commit adds a srcutree.convert_to_big option of zero that causes SRCU to decide at boot whether to wait for contention (small systems) or immediately expand to large (large systems). A new srcutree.big_cpu_lim (defaulting to 128) defines how many CPUs constitute a large system. Co-developed-by: Neeraj Upadhyay Signed-off-by: Neeraj Upadhyay Signed-off-by: Paul E. McKenney

srcu: Add contention-triggered addition of srcu_node tree

2022-04-11T22:52:30Z

This commit instruments the acquisitions of the srcu_struct structure's ->lock, enabling the initiation of a transition from SRCU_SIZE_SMALL to SRCU_SIZE_BIG when sufficient contention is experienced. The instrumentation counts the number of trylock failures within the confines of a single jiffy. If that number exceeds the value specified by the srcutree.small_contention_lim kernel boot parameter (which defaults to 100), and if the value specified by the srcutree.convert_to_big kernel boot parameter has the 0x10 bit set (defaults to 0), then a transition will be automatically initiated. By default, there will never be any transitions, so that none of the srcu_struct structures ever gains an srcu_node array. The useful values for srcutree.convert_to_big are: 0x00: Never convert. 0x01: Always convert at init_srcu_struct() time. 0x02: Convert when rcutorture prints its first round of statistics. 0x03: Decide conversion approach at boot given system size. 0x10: Convert if contention is encountered. 0x12: Convert if contention is encountered or when rcutorture prints its first round of statistics, whichever comes first. The value 0x11 acts the same as 0x01 because the conversion happens before there is any chance of contention. [ paulmck: Apply "static" feedback from kernel test robot. ] Co-developed-by: Neeraj Upadhyay Signed-off-by: Neeraj Upadhyay Signed-off-by: Paul E. McKenney

srcu: Create concurrency-safe helper for initiating size transition

2022-04-11T22:52:30Z

Once there are contention-initiated size transitions, it will be possible for rcutorture to initiate a transition at the same time as a contention-initiated transition. This commit therefore creates a concurrency-safe helper function named srcu_transition_to_big() to safely initiate size transitions. Co-developed-by: Neeraj Upadhyay Signed-off-by: Neeraj Upadhyay Signed-off-by: Paul E. McKenney

srcu: Explain srcu_funnel_gp_start() call to list_add() is safe

2022-04-11T22:52:30Z

This commit adds a comment explaining why an unprotected call to list_add() from srcu_funnel_gp_start() can be safe. TL;DR: It is only called during very early boot when we don't have no steeking concurrency! Co-developed-by: Neeraj Upadhyay Signed-off-by: Neeraj Upadhyay Signed-off-by: Paul E. McKenney

srcu: Prevent cleanup_srcu_struct() from freeing non-dynamic ->sda

2022-04-11T22:52:30Z

When an srcu_struct structure is created (but not in a kernel module) by DEFINE_SRCU() and friends, the per-CPU srcu_data structure is statically allocated. In all other cases, that structure is obtained from alloc_percpu(), in which case cleanup_srcu_struct() must invoke free_percpu() on the resulting ->sda pointer in the srcu_struct pointer. Which it does. Except that it also invokes free_percpu() on the ->sda pointer referencing the statically allocated per-CPU srcu_data structures. Which free_percpu() is surprisingly OK with. This commit nevertheless stops cleanup_srcu_struct() from freeing statically allocated per-CPU srcu_data structures. Co-developed-by: Neeraj Upadhyay Signed-off-by: Neeraj Upadhyay Signed-off-by: Paul E. McKenney