From f0f44752f5f61ee4e3bd88ae033fdb888320aafe Mon Sep 17 00:00:00 2001 From: Boqun Feng Date: Thu, 12 Jan 2023 22:59:54 -0800 Subject: rcu: Annotate SRCU's update-side lockdep dependencies Although all flavors of RCU readers are annotated correctly with lockdep as recursive read locks, they do not set the lock_acquire 'check' parameter. This means that RCU read locks are not added to the lockdep dependency graph, which in turn means that lockdep cannot detect RCU-based deadlocks. This is not a problem for RCU flavors having atomic read-side critical sections because context-based annotations can catch these deadlocks, see for example the RCU_LOCKDEP_WARN() statement in synchronize_rcu(). But context-based annotations are not helpful for sleepable RCU, especially given that it is perfectly legal to do synchronize_srcu(&srcu1) within an srcu_read_lock(&srcu2). However, we can detect SRCU-based by: (1) Making srcu_read_lock() a 'check'ed recursive read lock and (2) Making synchronize_srcu() a empty write lock critical section. Even better, with the newly introduced lock_sync(), we can avoid false positives about irq-unsafe/safe. This commit therefore makes it so. Note that NMI-safe SRCU read side critical sections are currently not annotated, but might be annotated in the future. Signed-off-by: Boqun Feng Signed-off-by: Paul E. McKenney [ boqun: Add comments for annotation per Waiman's suggestion ] [ boqun: Fix comment warning reported by Stephen Rothwell ] Acked-by: Peter Zijlstra (Intel) Signed-off-by: Boqun Feng --- kernel/rcu/srcutree.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index ab4ee58af84b..c541b82646b6 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -1307,6 +1307,8 @@ static void __synchronize_srcu(struct srcu_struct *ssp, bool do_norm) { struct rcu_synchronize rcu; + srcu_lock_sync(&ssp->dep_map); + RCU_LOCKDEP_WARN(lockdep_is_held(ssp) || lock_is_held(&rcu_bh_lock_map) || lock_is_held(&rcu_lock_map) || -- cgit From f4d01a259374ef358cd6b00a96b4dfc0fb05a844 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 13:28:04 -0700 Subject: srcu: Use static init for statically allocated in-module srcu_struct Further shrinking the srcu_struct structure is eased by requiring that in-module srcu_struct structures rely more heavily on static initialization. In particular, this preserves the property that a module-load-time srcu_struct initialization can fail only due to memory-allocation failure of the per-CPU srcu_data structures. It might also slightly improve robustness by keeping the number of memory allocations that must succeed down percpu_alloc() call. This is in preparation for splitting an srcu_usage structure out of the srcu_struct structure. [ paulmck: Fold in qiang1.zhang@intel.com feedback. ] Cc: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index ab4ee58af84b..7e6e7dfb1a87 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -1873,13 +1873,14 @@ void __init srcu_init(void) static int srcu_module_coming(struct module *mod) { int i; + struct srcu_struct *ssp; struct srcu_struct **sspp = mod->srcu_struct_ptrs; - int ret; for (i = 0; i < mod->num_srcu_structs; i++) { - ret = init_srcu_struct(*(sspp++)); - if (WARN_ON_ONCE(ret)) - return ret; + ssp = *(sspp++); + ssp->sda = alloc_percpu(struct srcu_data); + if (WARN_ON_ONCE(!ssp->sda)) + return -ENOMEM; } return 0; } @@ -1888,10 +1889,16 @@ static int srcu_module_coming(struct module *mod) static void srcu_module_going(struct module *mod) { int i; + struct srcu_struct *ssp; struct srcu_struct **sspp = mod->srcu_struct_ptrs; - for (i = 0; i < mod->num_srcu_structs; i++) - cleanup_srcu_struct(*(sspp++)); + for (i = 0; i < mod->num_srcu_structs; i++) { + ssp = *(sspp++); + if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_gp_seq_needed)) && + !WARN_ON_ONCE(!ssp->sda_is_static)) + cleanup_srcu_struct(ssp); + free_percpu(ssp->sda); + } } /* Handle one module, either coming or going. */ -- cgit From 95433f7263011e0e6e83caef85d98896dd99cab7 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Thu, 16 Mar 2023 17:58:51 -0700 Subject: srcu: Begin offloading srcu_struct fields to srcu_update MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The current srcu_struct structure is on the order of 200 bytes in size (depending on architecture and .config), which is much better than the old-style 26K bytes, but still all too inconvenient when one is trying to achieve good cache locality on a fastpath involving SRCU readers. However, only a few fields in srcu_struct are used by SRCU readers. The remaining fields could be offloaded to a new srcu_update structure, thus shrinking the srcu_struct structure down to a few tens of bytes. This commit begins this noble quest, a quest that is complicated by open-coded initialization of the srcu_struct within the srcu_notifier_head structure. This complication is addressed by updating the srcu_notifier_head structure's open coding, given that there does not appear to be a straightforward way of abstracting that initialization. This commit moves only the ->node pointer to srcu_update. Later commits will move additional fields. [ paulmck: Fold in qiang1.zhang@intel.com's memory-leak fix. ] Link: https://lore.kernel.org/all/20230320055751.4120251-1-qiang1.zhang@intel.com/ Suggested-by: Christoph Hellwig Cc: "Rafael J. Wysocki" Cc: "Michał Mirosław" Cc: Dmitry Osipenko Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Acked-by: Rafael J. Wysocki Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 28 +++++++++++++++++++--------- 1 file changed, 19 insertions(+), 9 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 7e6e7dfb1a87..049e20dbec76 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -173,12 +173,12 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) /* Initialize geometry if it has not already been initialized. */ rcu_init_geometry(); - ssp->node = kcalloc(rcu_num_nodes, sizeof(*ssp->node), gfp_flags); - if (!ssp->node) + ssp->srcu_sup->node = kcalloc(rcu_num_nodes, sizeof(*ssp->srcu_sup->node), gfp_flags); + if (!ssp->srcu_sup->node) return false; /* Work out the overall tree geometry. */ - ssp->level[0] = &ssp->node[0]; + ssp->level[0] = &ssp->srcu_sup->node[0]; for (i = 1; i < rcu_num_lvls; i++) ssp->level[i] = ssp->level[i - 1] + num_rcu_lvl[i - 1]; rcu_init_levelspread(levelspread, num_rcu_lvl); @@ -195,7 +195,7 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) snp->srcu_gp_seq_needed_exp = SRCU_SNP_INIT_SEQ; snp->grplo = -1; snp->grphi = -1; - if (snp == &ssp->node[0]) { + if (snp == &ssp->srcu_sup->node[0]) { /* Root node, special case. */ snp->srcu_parent = NULL; continue; @@ -236,8 +236,12 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) */ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) { + if (!is_static) + ssp->srcu_sup = kzalloc(sizeof(*ssp->srcu_sup), GFP_KERNEL); + if (!ssp->srcu_sup) + return -ENOMEM; ssp->srcu_size_state = SRCU_SIZE_SMALL; - ssp->node = NULL; + ssp->srcu_sup->node = NULL; mutex_init(&ssp->srcu_cb_mutex); mutex_init(&ssp->srcu_gp_mutex); ssp->srcu_idx = 0; @@ -249,8 +253,11 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) ssp->sda_is_static = is_static; if (!is_static) ssp->sda = alloc_percpu(struct srcu_data); - if (!ssp->sda) + if (!ssp->sda) { + if (!is_static) + kfree(ssp->srcu_sup); return -ENOMEM; + } init_srcu_struct_data(ssp); ssp->srcu_gp_seq_needed_exp = 0; ssp->srcu_last_gp_end = ktime_get_mono_fast_ns(); @@ -259,6 +266,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) if (!ssp->sda_is_static) { free_percpu(ssp->sda); ssp->sda = NULL; + kfree(ssp->srcu_sup); return -ENOMEM; } } else { @@ -656,13 +664,15 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) rcu_seq_current(&ssp->srcu_gp_seq), ssp->srcu_gp_seq_needed); return; /* Caller forgot to stop doing call_srcu()? */ } + kfree(ssp->srcu_sup->node); + ssp->srcu_sup->node = NULL; + ssp->srcu_size_state = SRCU_SIZE_SMALL; if (!ssp->sda_is_static) { free_percpu(ssp->sda); ssp->sda = NULL; + kfree(ssp->srcu_sup); + ssp->srcu_sup = NULL; } - kfree(ssp->node); - ssp->node = NULL; - ssp->srcu_size_state = SRCU_SIZE_SMALL; } EXPORT_SYMBOL_GPL(cleanup_srcu_struct); -- cgit From 208f41b1312443401353bec0c1939e2bfc28adce Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 14:43:08 -0700 Subject: srcu: Move ->level from srcu_struct to srcu_usage This commit moves the ->level[] array from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 049e20dbec76..acb0862faafa 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -178,9 +178,9 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) return false; /* Work out the overall tree geometry. */ - ssp->level[0] = &ssp->srcu_sup->node[0]; + ssp->srcu_sup->level[0] = &ssp->srcu_sup->node[0]; for (i = 1; i < rcu_num_lvls; i++) - ssp->level[i] = ssp->level[i - 1] + num_rcu_lvl[i - 1]; + ssp->srcu_sup->level[i] = ssp->srcu_sup->level[i - 1] + num_rcu_lvl[i - 1]; rcu_init_levelspread(levelspread, num_rcu_lvl); /* Each pass through this loop initializes one srcu_node structure. */ @@ -202,10 +202,10 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) } /* Non-root node. */ - if (snp == ssp->level[level + 1]) + if (snp == ssp->srcu_sup->level[level + 1]) level++; - snp->srcu_parent = ssp->level[level - 1] + - (snp - ssp->level[level]) / + snp->srcu_parent = ssp->srcu_sup->level[level - 1] + + (snp - ssp->srcu_sup->level[level]) / levelspread[level - 1]; } @@ -214,7 +214,7 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) * leaves of the srcu_node tree. */ level = rcu_num_lvls - 1; - snp_first = ssp->level[level]; + snp_first = ssp->srcu_sup->level[level]; for_each_possible_cpu(cpu) { sdp = per_cpu_ptr(ssp->sda, cpu); sdp->mynode = &snp_first[cpu / levelspread[level]]; @@ -889,7 +889,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) srcu_for_each_node_breadth_first(ssp, snp) { spin_lock_irq_rcu_node(snp); cbs = false; - last_lvl = snp >= ssp->level[rcu_num_lvls - 1]; + last_lvl = snp >= ssp->srcu_sup->level[rcu_num_lvls - 1]; if (last_lvl) cbs = ss_state < SRCU_SIZE_BIG || snp->srcu_have_cbs[idx] == gpseq; snp->srcu_have_cbs[idx] = gpseq; -- cgit From a0d8cbd3821369dc9478cabd605417afb9eb24dc Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 17:16:30 -0700 Subject: srcu: Move ->srcu_size_state from srcu_struct to srcu_usage This commit moves the ->srcu_size_state field from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 37 +++++++++++++++++++------------------ 1 file changed, 19 insertions(+), 18 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index acb0862faafa..8428a184d506 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -225,7 +225,7 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) } sdp->grpmask = 1 << (cpu - sdp->mynode->grplo); } - smp_store_release(&ssp->srcu_size_state, SRCU_SIZE_WAIT_BARRIER); + smp_store_release(&ssp->srcu_sup->srcu_size_state, SRCU_SIZE_WAIT_BARRIER); return true; } @@ -240,7 +240,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) ssp->srcu_sup = kzalloc(sizeof(*ssp->srcu_sup), GFP_KERNEL); if (!ssp->srcu_sup) return -ENOMEM; - ssp->srcu_size_state = SRCU_SIZE_SMALL; + ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; ssp->srcu_sup->node = NULL; mutex_init(&ssp->srcu_cb_mutex); mutex_init(&ssp->srcu_gp_mutex); @@ -261,7 +261,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) init_srcu_struct_data(ssp); ssp->srcu_gp_seq_needed_exp = 0; ssp->srcu_last_gp_end = ktime_get_mono_fast_ns(); - if (READ_ONCE(ssp->srcu_size_state) == SRCU_SIZE_SMALL && SRCU_SIZING_IS_INIT()) { + if (READ_ONCE(ssp->srcu_sup->srcu_size_state) == SRCU_SIZE_SMALL && SRCU_SIZING_IS_INIT()) { if (!init_srcu_struct_nodes(ssp, GFP_ATOMIC)) { if (!ssp->sda_is_static) { free_percpu(ssp->sda); @@ -270,7 +270,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) return -ENOMEM; } } else { - WRITE_ONCE(ssp->srcu_size_state, SRCU_SIZE_BIG); + WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_BIG); } } smp_store_release(&ssp->srcu_gp_seq_needed, 0); /* Init done. */ @@ -315,7 +315,7 @@ EXPORT_SYMBOL_GPL(init_srcu_struct); static void __srcu_transition_to_big(struct srcu_struct *ssp) { lockdep_assert_held(&ACCESS_PRIVATE(ssp, lock)); - smp_store_release(&ssp->srcu_size_state, SRCU_SIZE_ALLOC); + smp_store_release(&ssp->srcu_sup->srcu_size_state, SRCU_SIZE_ALLOC); } /* @@ -326,10 +326,10 @@ static void srcu_transition_to_big(struct srcu_struct *ssp) unsigned long flags; /* Double-checked locking on ->srcu_size-state. */ - if (smp_load_acquire(&ssp->srcu_size_state) != SRCU_SIZE_SMALL) + if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) return; spin_lock_irqsave_rcu_node(ssp, flags); - if (smp_load_acquire(&ssp->srcu_size_state) != SRCU_SIZE_SMALL) { + if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) { spin_unlock_irqrestore_rcu_node(ssp, flags); return; } @@ -345,7 +345,7 @@ static void spin_lock_irqsave_check_contention(struct srcu_struct *ssp) { unsigned long j; - if (!SRCU_SIZING_IS_CONTEND() || ssp->srcu_size_state) + if (!SRCU_SIZING_IS_CONTEND() || ssp->srcu_sup->srcu_size_state) return; j = jiffies; if (ssp->srcu_size_jiffies != j) { @@ -666,7 +666,7 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) } kfree(ssp->srcu_sup->node); ssp->srcu_sup->node = NULL; - ssp->srcu_size_state = SRCU_SIZE_SMALL; + ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; if (!ssp->sda_is_static) { free_percpu(ssp->sda); ssp->sda = NULL; @@ -770,7 +770,7 @@ static void srcu_gp_start(struct srcu_struct *ssp) struct srcu_data *sdp; int state; - if (smp_load_acquire(&ssp->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) + if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) sdp = per_cpu_ptr(ssp->sda, get_boot_cpu_id()); else sdp = this_cpu_ptr(ssp->sda); @@ -880,7 +880,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) /* A new grace period can start at this point. But only one. */ /* Initiate callback invocation as needed. */ - ss_state = smp_load_acquire(&ssp->srcu_size_state); + ss_state = smp_load_acquire(&ssp->srcu_sup->srcu_size_state); if (ss_state < SRCU_SIZE_WAIT_BARRIER) { srcu_schedule_cbs_sdp(per_cpu_ptr(ssp->sda, get_boot_cpu_id()), cbdelay); @@ -940,7 +940,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) if (ss_state == SRCU_SIZE_ALLOC) init_srcu_struct_nodes(ssp, GFP_KERNEL); else - smp_store_release(&ssp->srcu_size_state, ss_state + 1); + smp_store_release(&ssp->srcu_sup->srcu_size_state, ss_state + 1); } } @@ -1002,7 +1002,7 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, unsigned long snp_seq; /* Ensure that snp node tree is fully initialized before traversing it */ - if (smp_load_acquire(&ssp->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) + if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) snp_leaf = NULL; else snp_leaf = sdp->mynode; @@ -1209,7 +1209,7 @@ static unsigned long srcu_gp_start_if_needed(struct srcu_struct *ssp, * sequence number cannot wrap around in the meantime. */ idx = __srcu_read_lock_nmisafe(ssp); - ss_state = smp_load_acquire(&ssp->srcu_size_state); + ss_state = smp_load_acquire(&ssp->srcu_sup->srcu_size_state); if (ss_state < SRCU_SIZE_WAIT_CALL) sdp = per_cpu_ptr(ssp->sda, get_boot_cpu_id()); else @@ -1546,7 +1546,7 @@ void srcu_barrier(struct srcu_struct *ssp) atomic_set(&ssp->srcu_barrier_cpu_cnt, 1); idx = __srcu_read_lock_nmisafe(ssp); - if (smp_load_acquire(&ssp->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) + if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) srcu_barrier_one_cpu(ssp, per_cpu_ptr(ssp->sda, get_boot_cpu_id())); else for_each_possible_cpu(cpu) @@ -1784,7 +1784,7 @@ void srcu_torture_stats_print(struct srcu_struct *ssp, char *tt, char *tf) int cpu; int idx; unsigned long s0 = 0, s1 = 0; - int ss_state = READ_ONCE(ssp->srcu_size_state); + int ss_state = READ_ONCE(ssp->srcu_sup->srcu_size_state); int ss_state_idx = ss_state; idx = ssp->srcu_idx & 0x1; @@ -1871,8 +1871,9 @@ void __init srcu_init(void) ssp = list_first_entry(&srcu_boot_list, struct srcu_struct, work.work.entry); list_del_init(&ssp->work.work.entry); - if (SRCU_SIZING_IS(SRCU_SIZING_INIT) && ssp->srcu_size_state == SRCU_SIZE_SMALL) - ssp->srcu_size_state = SRCU_SIZE_ALLOC; + if (SRCU_SIZING_IS(SRCU_SIZING_INIT) && + ssp->srcu_sup->srcu_size_state == SRCU_SIZE_SMALL) + ssp->srcu_sup->srcu_size_state = SRCU_SIZE_ALLOC; queue_work(rcu_gp_wq, &ssp->work.work); } } -- cgit From 574dc1a7efe490dffe5c1ce0285306feec16a880 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 17:22:27 -0700 Subject: srcu: Move ->srcu_cb_mutex from srcu_struct to srcu_usage This commit moves the ->srcu_cb_mutex field from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 8428a184d506..1814f3bfc219 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -242,7 +242,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) return -ENOMEM; ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; ssp->srcu_sup->node = NULL; - mutex_init(&ssp->srcu_cb_mutex); + mutex_init(&ssp->srcu_sup->srcu_cb_mutex); mutex_init(&ssp->srcu_gp_mutex); ssp->srcu_idx = 0; ssp->srcu_gp_seq = 0; @@ -861,7 +861,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) int ss_state; /* Prevent more than one additional grace period. */ - mutex_lock(&ssp->srcu_cb_mutex); + mutex_lock(&ssp->srcu_sup->srcu_cb_mutex); /* End the current grace period. */ spin_lock_irq_rcu_node(ssp); @@ -921,7 +921,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) } /* Callback initiation done, allow grace periods after next. */ - mutex_unlock(&ssp->srcu_cb_mutex); + mutex_unlock(&ssp->srcu_sup->srcu_cb_mutex); /* Start a new grace period if needed. */ spin_lock_irq_rcu_node(ssp); -- cgit From 0839ade94bdef395bab02b3a579416c112062026 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 17:35:21 -0700 Subject: srcu: Move ->lock initialization after srcu_usage allocation Currently, both __init_srcu_struct() in CONFIG_DEBUG_LOCK_ALLOC=y kernels and init_srcu_struct() in CONFIG_DEBUG_LOCK_ALLOC=n kernel initialize the srcu_struct structure's ->lock before the srcu_usage structure has been allocated. This of course prevents the ->lock from being moved to the srcu_usage structure, so this commit moves the initialization into the init_srcu_struct_fields() after the srcu_usage structure has been allocated. Cc: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 1814f3bfc219..c2a024a60f1a 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -240,6 +240,8 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) ssp->srcu_sup = kzalloc(sizeof(*ssp->srcu_sup), GFP_KERNEL); if (!ssp->srcu_sup) return -ENOMEM; + if (!is_static) + spin_lock_init(&ACCESS_PRIVATE(ssp, lock)); ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; ssp->srcu_sup->node = NULL; mutex_init(&ssp->srcu_sup->srcu_cb_mutex); @@ -285,7 +287,6 @@ int __init_srcu_struct(struct srcu_struct *ssp, const char *name, /* Don't re-initialize a lock while it is held. */ debug_check_no_locks_freed((void *)ssp, sizeof(*ssp)); lockdep_init_map(&ssp->dep_map, name, key, 0); - spin_lock_init(&ACCESS_PRIVATE(ssp, lock)); return init_srcu_struct_fields(ssp, false); } EXPORT_SYMBOL_GPL(__init_srcu_struct); @@ -302,7 +303,6 @@ EXPORT_SYMBOL_GPL(__init_srcu_struct); */ int init_srcu_struct(struct srcu_struct *ssp) { - spin_lock_init(&ACCESS_PRIVATE(ssp, lock)); return init_srcu_struct_fields(ssp, false); } EXPORT_SYMBOL_GPL(init_srcu_struct); -- cgit From b3fb11f7e9c3c64dd86403409a070c996d8ac081 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 18:29:38 -0700 Subject: srcu: Move ->lock from srcu_struct to srcu_usage This commit moves the ->lock field from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 56 +++++++++++++++++++++++++-------------------------- 1 file changed, 28 insertions(+), 28 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index c2a024a60f1a..c42248cf18f6 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -103,7 +103,7 @@ do { \ #define spin_trylock_irqsave_rcu_node(p, flags) \ ({ \ - bool ___locked = spin_trylock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ + bool ___locked = spin_trylock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ \ if (___locked) \ smp_mb__after_unlock_lock(); \ @@ -241,7 +241,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) if (!ssp->srcu_sup) return -ENOMEM; if (!is_static) - spin_lock_init(&ACCESS_PRIVATE(ssp, lock)); + spin_lock_init(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; ssp->srcu_sup->node = NULL; mutex_init(&ssp->srcu_sup->srcu_cb_mutex); @@ -314,7 +314,7 @@ EXPORT_SYMBOL_GPL(init_srcu_struct); */ static void __srcu_transition_to_big(struct srcu_struct *ssp) { - lockdep_assert_held(&ACCESS_PRIVATE(ssp, lock)); + lockdep_assert_held(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); smp_store_release(&ssp->srcu_sup->srcu_size_state, SRCU_SIZE_ALLOC); } @@ -328,13 +328,13 @@ static void srcu_transition_to_big(struct srcu_struct *ssp) /* Double-checked locking on ->srcu_size-state. */ if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) return; - spin_lock_irqsave_rcu_node(ssp, flags); + spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) { - spin_unlock_irqrestore_rcu_node(ssp, flags); + spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); return; } __srcu_transition_to_big(ssp); - spin_unlock_irqrestore_rcu_node(ssp, flags); + spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -369,9 +369,9 @@ static void spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned lon if (spin_trylock_irqsave_rcu_node(sdp, *flags)) return; - spin_lock_irqsave_rcu_node(ssp, *flags); + spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); spin_lock_irqsave_check_contention(ssp); - spin_unlock_irqrestore_rcu_node(ssp, *flags); + spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, *flags); spin_lock_irqsave_rcu_node(sdp, *flags); } @@ -383,9 +383,9 @@ static void spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned lon */ static void spin_lock_irqsave_ssp_contention(struct srcu_struct *ssp, unsigned long *flags) { - if (spin_trylock_irqsave_rcu_node(ssp, *flags)) + if (spin_trylock_irqsave_rcu_node(ssp->srcu_sup, *flags)) return; - spin_lock_irqsave_rcu_node(ssp, *flags); + spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); spin_lock_irqsave_check_contention(ssp); } @@ -404,13 +404,13 @@ static void check_init_srcu_struct(struct srcu_struct *ssp) /* The smp_load_acquire() pairs with the smp_store_release(). */ if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_gp_seq_needed))) /*^^^*/ return; /* Already initialized. */ - spin_lock_irqsave_rcu_node(ssp, flags); + spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); if (!rcu_seq_state(ssp->srcu_gp_seq_needed)) { - spin_unlock_irqrestore_rcu_node(ssp, flags); + spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); return; } init_srcu_struct_fields(ssp, true); - spin_unlock_irqrestore_rcu_node(ssp, flags); + spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -774,7 +774,7 @@ static void srcu_gp_start(struct srcu_struct *ssp) sdp = per_cpu_ptr(ssp->sda, get_boot_cpu_id()); else sdp = this_cpu_ptr(ssp->sda); - lockdep_assert_held(&ACCESS_PRIVATE(ssp, lock)); + lockdep_assert_held(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); WARN_ON_ONCE(ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)); spin_lock_rcu_node(sdp); /* Interrupts already disabled. */ rcu_segcblist_advance(&sdp->srcu_cblist, @@ -864,7 +864,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) mutex_lock(&ssp->srcu_sup->srcu_cb_mutex); /* End the current grace period. */ - spin_lock_irq_rcu_node(ssp); + spin_lock_irq_rcu_node(ssp->srcu_sup); idx = rcu_seq_state(ssp->srcu_gp_seq); WARN_ON_ONCE(idx != SRCU_STATE_SCAN2); if (ULONG_CMP_LT(READ_ONCE(ssp->srcu_gp_seq), READ_ONCE(ssp->srcu_gp_seq_needed_exp))) @@ -875,7 +875,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) gpseq = rcu_seq_current(&ssp->srcu_gp_seq); if (ULONG_CMP_LT(ssp->srcu_gp_seq_needed_exp, gpseq)) WRITE_ONCE(ssp->srcu_gp_seq_needed_exp, gpseq); - spin_unlock_irq_rcu_node(ssp); + spin_unlock_irq_rcu_node(ssp->srcu_sup); mutex_unlock(&ssp->srcu_gp_mutex); /* A new grace period can start at this point. But only one. */ @@ -924,15 +924,15 @@ static void srcu_gp_end(struct srcu_struct *ssp) mutex_unlock(&ssp->srcu_sup->srcu_cb_mutex); /* Start a new grace period if needed. */ - spin_lock_irq_rcu_node(ssp); + spin_lock_irq_rcu_node(ssp->srcu_sup); gpseq = rcu_seq_current(&ssp->srcu_gp_seq); if (!rcu_seq_state(gpseq) && ULONG_CMP_LT(gpseq, ssp->srcu_gp_seq_needed)) { srcu_gp_start(ssp); - spin_unlock_irq_rcu_node(ssp); + spin_unlock_irq_rcu_node(ssp->srcu_sup); srcu_reschedule(ssp, 0); } else { - spin_unlock_irq_rcu_node(ssp); + spin_unlock_irq_rcu_node(ssp->srcu_sup); } /* Transition to big if needed. */ @@ -975,7 +975,7 @@ static void srcu_funnel_exp_start(struct srcu_struct *ssp, struct srcu_node *snp spin_lock_irqsave_ssp_contention(ssp, &flags); if (ULONG_CMP_LT(ssp->srcu_gp_seq_needed_exp, s)) WRITE_ONCE(ssp->srcu_gp_seq_needed_exp, s); - spin_unlock_irqrestore_rcu_node(ssp, flags); + spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -1064,7 +1064,7 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, else if (list_empty(&ssp->work.work.entry)) list_add(&ssp->work.work.entry, &srcu_boot_list); } - spin_unlock_irqrestore_rcu_node(ssp, flags); + spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -1599,17 +1599,17 @@ static void srcu_advance_state(struct srcu_struct *ssp) */ idx = rcu_seq_state(smp_load_acquire(&ssp->srcu_gp_seq)); /* ^^^ */ if (idx == SRCU_STATE_IDLE) { - spin_lock_irq_rcu_node(ssp); + spin_lock_irq_rcu_node(ssp->srcu_sup); if (ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)) { WARN_ON_ONCE(rcu_seq_state(ssp->srcu_gp_seq)); - spin_unlock_irq_rcu_node(ssp); + spin_unlock_irq_rcu_node(ssp->srcu_sup); mutex_unlock(&ssp->srcu_gp_mutex); return; } idx = rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)); if (idx == SRCU_STATE_IDLE) srcu_gp_start(ssp); - spin_unlock_irq_rcu_node(ssp); + spin_unlock_irq_rcu_node(ssp->srcu_sup); if (idx != SRCU_STATE_IDLE) { mutex_unlock(&ssp->srcu_gp_mutex); return; /* Someone else started the grace period. */ @@ -1623,10 +1623,10 @@ static void srcu_advance_state(struct srcu_struct *ssp) return; /* readers present, retry later. */ } srcu_flip(ssp); - spin_lock_irq_rcu_node(ssp); + spin_lock_irq_rcu_node(ssp->srcu_sup); rcu_seq_set_state(&ssp->srcu_gp_seq, SRCU_STATE_SCAN2); ssp->srcu_n_exp_nodelay = 0; - spin_unlock_irq_rcu_node(ssp); + spin_unlock_irq_rcu_node(ssp->srcu_sup); } if (rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)) == SRCU_STATE_SCAN2) { @@ -1710,7 +1710,7 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) { bool pushgp = true; - spin_lock_irq_rcu_node(ssp); + spin_lock_irq_rcu_node(ssp->srcu_sup); if (ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)) { if (!WARN_ON_ONCE(rcu_seq_state(ssp->srcu_gp_seq))) { /* All requests fulfilled, time to go idle. */ @@ -1720,7 +1720,7 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) /* Outstanding request and no GP. Start one. */ srcu_gp_start(ssp); } - spin_unlock_irq_rcu_node(ssp); + spin_unlock_irq_rcu_node(ssp->srcu_sup); if (pushgp) queue_delayed_work(rcu_gp_wq, &ssp->work, delay); -- cgit From e3a6ab25cfa0fcdcb31c346b9871a566d440980d Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 19:13:16 -0700 Subject: srcu: Move ->srcu_gp_mutex from srcu_struct to srcu_usage This commit moves the ->srcu_gp_mutex field from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index c42248cf18f6..a36066798de7 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -245,7 +245,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; ssp->srcu_sup->node = NULL; mutex_init(&ssp->srcu_sup->srcu_cb_mutex); - mutex_init(&ssp->srcu_gp_mutex); + mutex_init(&ssp->srcu_sup->srcu_gp_mutex); ssp->srcu_idx = 0; ssp->srcu_gp_seq = 0; ssp->srcu_barrier_seq = 0; @@ -876,7 +876,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) if (ULONG_CMP_LT(ssp->srcu_gp_seq_needed_exp, gpseq)) WRITE_ONCE(ssp->srcu_gp_seq_needed_exp, gpseq); spin_unlock_irq_rcu_node(ssp->srcu_sup); - mutex_unlock(&ssp->srcu_gp_mutex); + mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); /* A new grace period can start at this point. But only one. */ /* Initiate callback invocation as needed. */ @@ -1585,7 +1585,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) { int idx; - mutex_lock(&ssp->srcu_gp_mutex); + mutex_lock(&ssp->srcu_sup->srcu_gp_mutex); /* * Because readers might be delayed for an extended period after @@ -1603,7 +1603,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) if (ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)) { WARN_ON_ONCE(rcu_seq_state(ssp->srcu_gp_seq)); spin_unlock_irq_rcu_node(ssp->srcu_sup); - mutex_unlock(&ssp->srcu_gp_mutex); + mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; } idx = rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)); @@ -1611,7 +1611,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) srcu_gp_start(ssp); spin_unlock_irq_rcu_node(ssp->srcu_sup); if (idx != SRCU_STATE_IDLE) { - mutex_unlock(&ssp->srcu_gp_mutex); + mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* Someone else started the grace period. */ } } @@ -1619,7 +1619,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) if (rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)) == SRCU_STATE_SCAN1) { idx = 1 ^ (ssp->srcu_idx & 1); if (!try_check_zero(ssp, idx, 1)) { - mutex_unlock(&ssp->srcu_gp_mutex); + mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* readers present, retry later. */ } srcu_flip(ssp); @@ -1637,7 +1637,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) */ idx = 1 ^ (ssp->srcu_idx & 1); if (!try_check_zero(ssp, idx, 2)) { - mutex_unlock(&ssp->srcu_gp_mutex); + mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* readers present, retry later. */ } ssp->srcu_n_exp_nodelay = 0; -- cgit From 03200b5ca3b4d4edf634dc052bf3b8eb8dc8bbbc Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 19:30:50 -0700 Subject: srcu: Move grace-period fields from srcu_struct to srcu_usage This commit moves the ->srcu_gp_seq, ->srcu_gp_seq_needed, ->srcu_gp_seq_needed_exp, ->srcu_gp_start, and ->srcu_last_gp_end fields from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 128 +++++++++++++++++++++++++------------------------- 1 file changed, 64 insertions(+), 64 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index a36066798de7..340eb685cf64 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -135,8 +135,8 @@ static void init_srcu_struct_data(struct srcu_struct *ssp) spin_lock_init(&ACCESS_PRIVATE(sdp, lock)); rcu_segcblist_init(&sdp->srcu_cblist); sdp->srcu_cblist_invoking = false; - sdp->srcu_gp_seq_needed = ssp->srcu_gp_seq; - sdp->srcu_gp_seq_needed_exp = ssp->srcu_gp_seq; + sdp->srcu_gp_seq_needed = ssp->srcu_sup->srcu_gp_seq; + sdp->srcu_gp_seq_needed_exp = ssp->srcu_sup->srcu_gp_seq; sdp->mynode = NULL; sdp->cpu = cpu; INIT_WORK(&sdp->work, srcu_invoke_callbacks); @@ -247,7 +247,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) mutex_init(&ssp->srcu_sup->srcu_cb_mutex); mutex_init(&ssp->srcu_sup->srcu_gp_mutex); ssp->srcu_idx = 0; - ssp->srcu_gp_seq = 0; + ssp->srcu_sup->srcu_gp_seq = 0; ssp->srcu_barrier_seq = 0; mutex_init(&ssp->srcu_barrier_mutex); atomic_set(&ssp->srcu_barrier_cpu_cnt, 0); @@ -261,8 +261,8 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) return -ENOMEM; } init_srcu_struct_data(ssp); - ssp->srcu_gp_seq_needed_exp = 0; - ssp->srcu_last_gp_end = ktime_get_mono_fast_ns(); + ssp->srcu_sup->srcu_gp_seq_needed_exp = 0; + ssp->srcu_sup->srcu_last_gp_end = ktime_get_mono_fast_ns(); if (READ_ONCE(ssp->srcu_sup->srcu_size_state) == SRCU_SIZE_SMALL && SRCU_SIZING_IS_INIT()) { if (!init_srcu_struct_nodes(ssp, GFP_ATOMIC)) { if (!ssp->sda_is_static) { @@ -275,7 +275,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_BIG); } } - smp_store_release(&ssp->srcu_gp_seq_needed, 0); /* Init done. */ + smp_store_release(&ssp->srcu_sup->srcu_gp_seq_needed, 0); /* Init done. */ return 0; } @@ -402,10 +402,10 @@ static void check_init_srcu_struct(struct srcu_struct *ssp) unsigned long flags; /* The smp_load_acquire() pairs with the smp_store_release(). */ - if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_gp_seq_needed))) /*^^^*/ + if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq_needed))) /*^^^*/ return; /* Already initialized. */ spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); - if (!rcu_seq_state(ssp->srcu_gp_seq_needed)) { + if (!rcu_seq_state(ssp->srcu_sup->srcu_gp_seq_needed)) { spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); return; } @@ -616,11 +616,11 @@ static unsigned long srcu_get_delay(struct srcu_struct *ssp) unsigned long j; unsigned long jbase = SRCU_INTERVAL; - if (ULONG_CMP_LT(READ_ONCE(ssp->srcu_gp_seq), READ_ONCE(ssp->srcu_gp_seq_needed_exp))) + if (ULONG_CMP_LT(READ_ONCE(ssp->srcu_sup->srcu_gp_seq), READ_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp))) jbase = 0; - if (rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq))) { + if (rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq))) { j = jiffies - 1; - gpstart = READ_ONCE(ssp->srcu_gp_start); + gpstart = READ_ONCE(ssp->srcu_sup->srcu_gp_start); if (time_after(j, gpstart)) jbase += j - gpstart; if (!jbase) { @@ -656,12 +656,12 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) if (WARN_ON(rcu_segcblist_n_cbs(&sdp->srcu_cblist))) return; /* Forgot srcu_barrier(), so just leak it! */ } - if (WARN_ON(rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)) != SRCU_STATE_IDLE) || - WARN_ON(rcu_seq_current(&ssp->srcu_gp_seq) != ssp->srcu_gp_seq_needed) || + if (WARN_ON(rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)) != SRCU_STATE_IDLE) || + WARN_ON(rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq) != ssp->srcu_sup->srcu_gp_seq_needed) || WARN_ON(srcu_readers_active(ssp))) { pr_info("%s: Active srcu_struct %p read state: %d gp state: %lu/%lu\n", - __func__, ssp, rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)), - rcu_seq_current(&ssp->srcu_gp_seq), ssp->srcu_gp_seq_needed); + __func__, ssp, rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)), + rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq), ssp->srcu_sup->srcu_gp_seq_needed); return; /* Caller forgot to stop doing call_srcu()? */ } kfree(ssp->srcu_sup->node); @@ -775,18 +775,18 @@ static void srcu_gp_start(struct srcu_struct *ssp) else sdp = this_cpu_ptr(ssp->sda); lockdep_assert_held(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); - WARN_ON_ONCE(ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)); + WARN_ON_ONCE(ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)); spin_lock_rcu_node(sdp); /* Interrupts already disabled. */ rcu_segcblist_advance(&sdp->srcu_cblist, - rcu_seq_current(&ssp->srcu_gp_seq)); + rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq)); (void)rcu_segcblist_accelerate(&sdp->srcu_cblist, - rcu_seq_snap(&ssp->srcu_gp_seq)); + rcu_seq_snap(&ssp->srcu_sup->srcu_gp_seq)); spin_unlock_rcu_node(sdp); /* Interrupts remain disabled. */ - WRITE_ONCE(ssp->srcu_gp_start, jiffies); + WRITE_ONCE(ssp->srcu_sup->srcu_gp_start, jiffies); WRITE_ONCE(ssp->srcu_n_exp_nodelay, 0); smp_mb(); /* Order prior store to ->srcu_gp_seq_needed vs. GP start. */ - rcu_seq_start(&ssp->srcu_gp_seq); - state = rcu_seq_state(ssp->srcu_gp_seq); + rcu_seq_start(&ssp->srcu_sup->srcu_gp_seq); + state = rcu_seq_state(ssp->srcu_sup->srcu_gp_seq); WARN_ON_ONCE(state != SRCU_STATE_SCAN1); } @@ -865,16 +865,16 @@ static void srcu_gp_end(struct srcu_struct *ssp) /* End the current grace period. */ spin_lock_irq_rcu_node(ssp->srcu_sup); - idx = rcu_seq_state(ssp->srcu_gp_seq); + idx = rcu_seq_state(ssp->srcu_sup->srcu_gp_seq); WARN_ON_ONCE(idx != SRCU_STATE_SCAN2); - if (ULONG_CMP_LT(READ_ONCE(ssp->srcu_gp_seq), READ_ONCE(ssp->srcu_gp_seq_needed_exp))) + if (ULONG_CMP_LT(READ_ONCE(ssp->srcu_sup->srcu_gp_seq), READ_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp))) cbdelay = 0; - WRITE_ONCE(ssp->srcu_last_gp_end, ktime_get_mono_fast_ns()); - rcu_seq_end(&ssp->srcu_gp_seq); - gpseq = rcu_seq_current(&ssp->srcu_gp_seq); - if (ULONG_CMP_LT(ssp->srcu_gp_seq_needed_exp, gpseq)) - WRITE_ONCE(ssp->srcu_gp_seq_needed_exp, gpseq); + WRITE_ONCE(ssp->srcu_sup->srcu_last_gp_end, ktime_get_mono_fast_ns()); + rcu_seq_end(&ssp->srcu_sup->srcu_gp_seq); + gpseq = rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq); + if (ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed_exp, gpseq)) + WRITE_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp, gpseq); spin_unlock_irq_rcu_node(ssp->srcu_sup); mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); /* A new grace period can start at this point. But only one. */ @@ -925,9 +925,9 @@ static void srcu_gp_end(struct srcu_struct *ssp) /* Start a new grace period if needed. */ spin_lock_irq_rcu_node(ssp->srcu_sup); - gpseq = rcu_seq_current(&ssp->srcu_gp_seq); + gpseq = rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq); if (!rcu_seq_state(gpseq) && - ULONG_CMP_LT(gpseq, ssp->srcu_gp_seq_needed)) { + ULONG_CMP_LT(gpseq, ssp->srcu_sup->srcu_gp_seq_needed)) { srcu_gp_start(ssp); spin_unlock_irq_rcu_node(ssp->srcu_sup); srcu_reschedule(ssp, 0); @@ -960,7 +960,7 @@ static void srcu_funnel_exp_start(struct srcu_struct *ssp, struct srcu_node *snp if (snp) for (; snp != NULL; snp = snp->srcu_parent) { sgsne = READ_ONCE(snp->srcu_gp_seq_needed_exp); - if (WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_gp_seq, s)) || + if (WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, s)) || (!srcu_invl_snp_seq(sgsne) && ULONG_CMP_GE(sgsne, s))) return; spin_lock_irqsave_rcu_node(snp, flags); @@ -973,8 +973,8 @@ static void srcu_funnel_exp_start(struct srcu_struct *ssp, struct srcu_node *snp spin_unlock_irqrestore_rcu_node(snp, flags); } spin_lock_irqsave_ssp_contention(ssp, &flags); - if (ULONG_CMP_LT(ssp->srcu_gp_seq_needed_exp, s)) - WRITE_ONCE(ssp->srcu_gp_seq_needed_exp, s); + if (ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed_exp, s)) + WRITE_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp, s); spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } @@ -1010,7 +1010,7 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, if (snp_leaf) /* Each pass through the loop does one level of the srcu_node tree. */ for (snp = snp_leaf; snp != NULL; snp = snp->srcu_parent) { - if (WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_gp_seq, s)) && snp != snp_leaf) + if (WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, s)) && snp != snp_leaf) return; /* GP already done and CBs recorded. */ spin_lock_irqsave_rcu_node(snp, flags); snp_seq = snp->srcu_have_cbs[idx]; @@ -1037,20 +1037,20 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, /* Top of tree, must ensure the grace period will be started. */ spin_lock_irqsave_ssp_contention(ssp, &flags); - if (ULONG_CMP_LT(ssp->srcu_gp_seq_needed, s)) { + if (ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed, s)) { /* * Record need for grace period s. Pair with load * acquire setting up for initialization. */ - smp_store_release(&ssp->srcu_gp_seq_needed, s); /*^^^*/ + smp_store_release(&ssp->srcu_sup->srcu_gp_seq_needed, s); /*^^^*/ } - if (!do_norm && ULONG_CMP_LT(ssp->srcu_gp_seq_needed_exp, s)) - WRITE_ONCE(ssp->srcu_gp_seq_needed_exp, s); + if (!do_norm && ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed_exp, s)) + WRITE_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp, s); /* If grace period not already in progress, start it. */ - if (!WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_gp_seq, s)) && - rcu_seq_state(ssp->srcu_gp_seq) == SRCU_STATE_IDLE) { - WARN_ON_ONCE(ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)); + if (!WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, s)) && + rcu_seq_state(ssp->srcu_sup->srcu_gp_seq) == SRCU_STATE_IDLE) { + WARN_ON_ONCE(ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)); srcu_gp_start(ssp); // And how can that list_add() in the "else" clause @@ -1164,18 +1164,18 @@ static bool srcu_might_be_idle(struct srcu_struct *ssp) /* First, see if enough time has passed since the last GP. */ t = ktime_get_mono_fast_ns(); - tlast = READ_ONCE(ssp->srcu_last_gp_end); + tlast = READ_ONCE(ssp->srcu_sup->srcu_last_gp_end); if (exp_holdoff == 0 || time_in_range_open(t, tlast, tlast + exp_holdoff)) return false; /* Too soon after last GP. */ /* Next, check for probable idleness. */ - curseq = rcu_seq_current(&ssp->srcu_gp_seq); + curseq = rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq); smp_mb(); /* Order ->srcu_gp_seq with ->srcu_gp_seq_needed. */ - if (ULONG_CMP_LT(curseq, READ_ONCE(ssp->srcu_gp_seq_needed))) + if (ULONG_CMP_LT(curseq, READ_ONCE(ssp->srcu_sup->srcu_gp_seq_needed))) return false; /* Grace period in progress, so not idle. */ smp_mb(); /* Order ->srcu_gp_seq with prior access. */ - if (curseq != rcu_seq_current(&ssp->srcu_gp_seq)) + if (curseq != rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq)) return false; /* GP # changed, so not idle. */ return true; /* With reasonable probability, idle! */ } @@ -1218,8 +1218,8 @@ static unsigned long srcu_gp_start_if_needed(struct srcu_struct *ssp, if (rhp) rcu_segcblist_enqueue(&sdp->srcu_cblist, rhp); rcu_segcblist_advance(&sdp->srcu_cblist, - rcu_seq_current(&ssp->srcu_gp_seq)); - s = rcu_seq_snap(&ssp->srcu_gp_seq); + rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq)); + s = rcu_seq_snap(&ssp->srcu_sup->srcu_gp_seq); (void)rcu_segcblist_accelerate(&sdp->srcu_cblist, s); if (ULONG_CMP_LT(sdp->srcu_gp_seq_needed, s)) { sdp->srcu_gp_seq_needed = s; @@ -1430,7 +1430,7 @@ unsigned long get_state_synchronize_srcu(struct srcu_struct *ssp) // Any prior manipulation of SRCU-protected data must happen // before the load from ->srcu_gp_seq. smp_mb(); - return rcu_seq_snap(&ssp->srcu_gp_seq); + return rcu_seq_snap(&ssp->srcu_sup->srcu_gp_seq); } EXPORT_SYMBOL_GPL(get_state_synchronize_srcu); @@ -1477,7 +1477,7 @@ EXPORT_SYMBOL_GPL(start_poll_synchronize_srcu); */ bool poll_state_synchronize_srcu(struct srcu_struct *ssp, unsigned long cookie) { - if (!rcu_seq_done(&ssp->srcu_gp_seq, cookie)) + if (!rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, cookie)) return false; // Ensure that the end of the SRCU grace period happens before // any subsequent code that the caller might execute. @@ -1597,16 +1597,16 @@ static void srcu_advance_state(struct srcu_struct *ssp) * The load-acquire ensures that we see the accesses performed * by the prior grace period. */ - idx = rcu_seq_state(smp_load_acquire(&ssp->srcu_gp_seq)); /* ^^^ */ + idx = rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq)); /* ^^^ */ if (idx == SRCU_STATE_IDLE) { spin_lock_irq_rcu_node(ssp->srcu_sup); - if (ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)) { - WARN_ON_ONCE(rcu_seq_state(ssp->srcu_gp_seq)); + if (ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)) { + WARN_ON_ONCE(rcu_seq_state(ssp->srcu_sup->srcu_gp_seq)); spin_unlock_irq_rcu_node(ssp->srcu_sup); mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; } - idx = rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)); + idx = rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)); if (idx == SRCU_STATE_IDLE) srcu_gp_start(ssp); spin_unlock_irq_rcu_node(ssp->srcu_sup); @@ -1616,7 +1616,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) } } - if (rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)) == SRCU_STATE_SCAN1) { + if (rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)) == SRCU_STATE_SCAN1) { idx = 1 ^ (ssp->srcu_idx & 1); if (!try_check_zero(ssp, idx, 1)) { mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); @@ -1624,12 +1624,12 @@ static void srcu_advance_state(struct srcu_struct *ssp) } srcu_flip(ssp); spin_lock_irq_rcu_node(ssp->srcu_sup); - rcu_seq_set_state(&ssp->srcu_gp_seq, SRCU_STATE_SCAN2); + rcu_seq_set_state(&ssp->srcu_sup->srcu_gp_seq, SRCU_STATE_SCAN2); ssp->srcu_n_exp_nodelay = 0; spin_unlock_irq_rcu_node(ssp->srcu_sup); } - if (rcu_seq_state(READ_ONCE(ssp->srcu_gp_seq)) == SRCU_STATE_SCAN2) { + if (rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)) == SRCU_STATE_SCAN2) { /* * SRCU read-side critical sections are normally short, @@ -1666,7 +1666,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) rcu_cblist_init(&ready_cbs); spin_lock_irq_rcu_node(sdp); rcu_segcblist_advance(&sdp->srcu_cblist, - rcu_seq_current(&ssp->srcu_gp_seq)); + rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq)); if (sdp->srcu_cblist_invoking || !rcu_segcblist_ready_cbs(&sdp->srcu_cblist)) { spin_unlock_irq_rcu_node(sdp); @@ -1694,7 +1694,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) spin_lock_irq_rcu_node(sdp); rcu_segcblist_add_len(&sdp->srcu_cblist, -len); (void)rcu_segcblist_accelerate(&sdp->srcu_cblist, - rcu_seq_snap(&ssp->srcu_gp_seq)); + rcu_seq_snap(&ssp->srcu_sup->srcu_gp_seq)); sdp->srcu_cblist_invoking = false; more = rcu_segcblist_ready_cbs(&sdp->srcu_cblist); spin_unlock_irq_rcu_node(sdp); @@ -1711,12 +1711,12 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) bool pushgp = true; spin_lock_irq_rcu_node(ssp->srcu_sup); - if (ULONG_CMP_GE(ssp->srcu_gp_seq, ssp->srcu_gp_seq_needed)) { - if (!WARN_ON_ONCE(rcu_seq_state(ssp->srcu_gp_seq))) { + if (ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)) { + if (!WARN_ON_ONCE(rcu_seq_state(ssp->srcu_sup->srcu_gp_seq))) { /* All requests fulfilled, time to go idle. */ pushgp = false; } - } else if (!rcu_seq_state(ssp->srcu_gp_seq)) { + } else if (!rcu_seq_state(ssp->srcu_sup->srcu_gp_seq)) { /* Outstanding request and no GP. Start one. */ srcu_gp_start(ssp); } @@ -1762,7 +1762,7 @@ void srcutorture_get_gp_data(enum rcutorture_type test_type, if (test_type != SRCU_FLAVOR) return; *flags = 0; - *gp_seq = rcu_seq_current(&ssp->srcu_gp_seq); + *gp_seq = rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq); } EXPORT_SYMBOL_GPL(srcutorture_get_gp_data); @@ -1791,7 +1791,7 @@ void srcu_torture_stats_print(struct srcu_struct *ssp, char *tt, char *tf) if (ss_state < 0 || ss_state >= ARRAY_SIZE(srcu_size_state_name)) ss_state_idx = ARRAY_SIZE(srcu_size_state_name) - 1; pr_alert("%s%s Tree SRCU g%ld state %d (%s)", - tt, tf, rcu_seq_current(&ssp->srcu_gp_seq), ss_state, + tt, tf, rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq), ss_state, srcu_size_state_name[ss_state_idx]); if (!ssp->sda) { // Called after cleanup_srcu_struct(), perhaps. @@ -1905,7 +1905,7 @@ static void srcu_module_going(struct module *mod) for (i = 0; i < mod->num_srcu_structs; i++) { ssp = *(sspp++); - if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_gp_seq_needed)) && + if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq_needed)) && !WARN_ON_ONCE(!ssp->sda_is_static)) cleanup_srcu_struct(ssp); free_percpu(ssp->sda); -- cgit From 3b46679c623c2766f4c56fd3f9ce8edbb38c5d20 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 20:01:02 -0700 Subject: srcu: Move heuristics fields from srcu_struct to srcu_usage This commit moves the ->srcu_size_jiffies, ->srcu_n_lock_retries, and ->srcu_n_exp_nodelay fields from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 340eb685cf64..291fb520bce0 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -348,11 +348,11 @@ static void spin_lock_irqsave_check_contention(struct srcu_struct *ssp) if (!SRCU_SIZING_IS_CONTEND() || ssp->srcu_sup->srcu_size_state) return; j = jiffies; - if (ssp->srcu_size_jiffies != j) { - ssp->srcu_size_jiffies = j; - ssp->srcu_n_lock_retries = 0; + if (ssp->srcu_sup->srcu_size_jiffies != j) { + ssp->srcu_sup->srcu_size_jiffies = j; + ssp->srcu_sup->srcu_n_lock_retries = 0; } - if (++ssp->srcu_n_lock_retries <= small_contention_lim) + if (++ssp->srcu_sup->srcu_n_lock_retries <= small_contention_lim) return; __srcu_transition_to_big(ssp); } @@ -624,8 +624,8 @@ static unsigned long srcu_get_delay(struct srcu_struct *ssp) if (time_after(j, gpstart)) jbase += j - gpstart; if (!jbase) { - WRITE_ONCE(ssp->srcu_n_exp_nodelay, READ_ONCE(ssp->srcu_n_exp_nodelay) + 1); - if (READ_ONCE(ssp->srcu_n_exp_nodelay) > srcu_max_nodelay_phase) + WRITE_ONCE(ssp->srcu_sup->srcu_n_exp_nodelay, READ_ONCE(ssp->srcu_sup->srcu_n_exp_nodelay) + 1); + if (READ_ONCE(ssp->srcu_sup->srcu_n_exp_nodelay) > srcu_max_nodelay_phase) jbase = 1; } } @@ -783,7 +783,7 @@ static void srcu_gp_start(struct srcu_struct *ssp) rcu_seq_snap(&ssp->srcu_sup->srcu_gp_seq)); spin_unlock_rcu_node(sdp); /* Interrupts remain disabled. */ WRITE_ONCE(ssp->srcu_sup->srcu_gp_start, jiffies); - WRITE_ONCE(ssp->srcu_n_exp_nodelay, 0); + WRITE_ONCE(ssp->srcu_sup->srcu_n_exp_nodelay, 0); smp_mb(); /* Order prior store to ->srcu_gp_seq_needed vs. GP start. */ rcu_seq_start(&ssp->srcu_sup->srcu_gp_seq); state = rcu_seq_state(ssp->srcu_sup->srcu_gp_seq); @@ -1625,7 +1625,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) srcu_flip(ssp); spin_lock_irq_rcu_node(ssp->srcu_sup); rcu_seq_set_state(&ssp->srcu_sup->srcu_gp_seq, SRCU_STATE_SCAN2); - ssp->srcu_n_exp_nodelay = 0; + ssp->srcu_sup->srcu_n_exp_nodelay = 0; spin_unlock_irq_rcu_node(ssp->srcu_sup); } @@ -1640,7 +1640,7 @@ static void srcu_advance_state(struct srcu_struct *ssp) mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* readers present, retry later. */ } - ssp->srcu_n_exp_nodelay = 0; + ssp->srcu_sup->srcu_n_exp_nodelay = 0; srcu_gp_end(ssp); /* Releases ->srcu_gp_mutex. */ } } -- cgit From 660349ac79cb22bb64c44b026d879069783e97d5 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 20:22:58 -0700 Subject: srcu: Move ->sda_is_static from srcu_struct to srcu_usage This commit moves the ->sda_is_static field from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 291fb520bce0..20f2373f7e25 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -252,7 +252,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) mutex_init(&ssp->srcu_barrier_mutex); atomic_set(&ssp->srcu_barrier_cpu_cnt, 0); INIT_DELAYED_WORK(&ssp->work, process_srcu); - ssp->sda_is_static = is_static; + ssp->srcu_sup->sda_is_static = is_static; if (!is_static) ssp->sda = alloc_percpu(struct srcu_data); if (!ssp->sda) { @@ -265,7 +265,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) ssp->srcu_sup->srcu_last_gp_end = ktime_get_mono_fast_ns(); if (READ_ONCE(ssp->srcu_sup->srcu_size_state) == SRCU_SIZE_SMALL && SRCU_SIZING_IS_INIT()) { if (!init_srcu_struct_nodes(ssp, GFP_ATOMIC)) { - if (!ssp->sda_is_static) { + if (!ssp->srcu_sup->sda_is_static) { free_percpu(ssp->sda); ssp->sda = NULL; kfree(ssp->srcu_sup); @@ -667,7 +667,7 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) kfree(ssp->srcu_sup->node); ssp->srcu_sup->node = NULL; ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; - if (!ssp->sda_is_static) { + if (!ssp->srcu_sup->sda_is_static) { free_percpu(ssp->sda); ssp->sda = NULL; kfree(ssp->srcu_sup); @@ -1906,7 +1906,7 @@ static void srcu_module_going(struct module *mod) for (i = 0; i < mod->num_srcu_structs; i++) { ssp = *(sspp++); if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq_needed)) && - !WARN_ON_ONCE(!ssp->sda_is_static)) + !WARN_ON_ONCE(!ssp->srcu_sup->sda_is_static)) cleanup_srcu_struct(ssp); free_percpu(ssp->sda); } -- cgit From d20162e0bfc222183a7c94cd00e74b6bbf1a605b Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 21:08:18 -0700 Subject: srcu: Move srcu_barrier() fields from srcu_struct to srcu_usage This commit moves the ->srcu_barrier_seq, ->srcu_barrier_mutex, ->srcu_barrier_completion, and ->srcu_barrier_cpu_cnt fields from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 20f2373f7e25..97d1fe9a160c 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -248,9 +248,9 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) mutex_init(&ssp->srcu_sup->srcu_gp_mutex); ssp->srcu_idx = 0; ssp->srcu_sup->srcu_gp_seq = 0; - ssp->srcu_barrier_seq = 0; - mutex_init(&ssp->srcu_barrier_mutex); - atomic_set(&ssp->srcu_barrier_cpu_cnt, 0); + ssp->srcu_sup->srcu_barrier_seq = 0; + mutex_init(&ssp->srcu_sup->srcu_barrier_mutex); + atomic_set(&ssp->srcu_sup->srcu_barrier_cpu_cnt, 0); INIT_DELAYED_WORK(&ssp->work, process_srcu); ssp->srcu_sup->sda_is_static = is_static; if (!is_static) @@ -1496,8 +1496,8 @@ static void srcu_barrier_cb(struct rcu_head *rhp) sdp = container_of(rhp, struct srcu_data, srcu_barrier_head); ssp = sdp->ssp; - if (atomic_dec_and_test(&ssp->srcu_barrier_cpu_cnt)) - complete(&ssp->srcu_barrier_completion); + if (atomic_dec_and_test(&ssp->srcu_sup->srcu_barrier_cpu_cnt)) + complete(&ssp->srcu_sup->srcu_barrier_completion); } /* @@ -1511,13 +1511,13 @@ static void srcu_barrier_cb(struct rcu_head *rhp) static void srcu_barrier_one_cpu(struct srcu_struct *ssp, struct srcu_data *sdp) { spin_lock_irq_rcu_node(sdp); - atomic_inc(&ssp->srcu_barrier_cpu_cnt); + atomic_inc(&ssp->srcu_sup->srcu_barrier_cpu_cnt); sdp->srcu_barrier_head.func = srcu_barrier_cb; debug_rcu_head_queue(&sdp->srcu_barrier_head); if (!rcu_segcblist_entrain(&sdp->srcu_cblist, &sdp->srcu_barrier_head)) { debug_rcu_head_unqueue(&sdp->srcu_barrier_head); - atomic_dec(&ssp->srcu_barrier_cpu_cnt); + atomic_dec(&ssp->srcu_sup->srcu_barrier_cpu_cnt); } spin_unlock_irq_rcu_node(sdp); } @@ -1530,20 +1530,20 @@ void srcu_barrier(struct srcu_struct *ssp) { int cpu; int idx; - unsigned long s = rcu_seq_snap(&ssp->srcu_barrier_seq); + unsigned long s = rcu_seq_snap(&ssp->srcu_sup->srcu_barrier_seq); check_init_srcu_struct(ssp); - mutex_lock(&ssp->srcu_barrier_mutex); - if (rcu_seq_done(&ssp->srcu_barrier_seq, s)) { + mutex_lock(&ssp->srcu_sup->srcu_barrier_mutex); + if (rcu_seq_done(&ssp->srcu_sup->srcu_barrier_seq, s)) { smp_mb(); /* Force ordering following return. */ - mutex_unlock(&ssp->srcu_barrier_mutex); + mutex_unlock(&ssp->srcu_sup->srcu_barrier_mutex); return; /* Someone else did our work for us. */ } - rcu_seq_start(&ssp->srcu_barrier_seq); - init_completion(&ssp->srcu_barrier_completion); + rcu_seq_start(&ssp->srcu_sup->srcu_barrier_seq); + init_completion(&ssp->srcu_sup->srcu_barrier_completion); /* Initial count prevents reaching zero until all CBs are posted. */ - atomic_set(&ssp->srcu_barrier_cpu_cnt, 1); + atomic_set(&ssp->srcu_sup->srcu_barrier_cpu_cnt, 1); idx = __srcu_read_lock_nmisafe(ssp); if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) @@ -1554,12 +1554,12 @@ void srcu_barrier(struct srcu_struct *ssp) __srcu_read_unlock_nmisafe(ssp, idx); /* Remove the initial count, at which point reaching zero can happen. */ - if (atomic_dec_and_test(&ssp->srcu_barrier_cpu_cnt)) - complete(&ssp->srcu_barrier_completion); - wait_for_completion(&ssp->srcu_barrier_completion); + if (atomic_dec_and_test(&ssp->srcu_sup->srcu_barrier_cpu_cnt)) + complete(&ssp->srcu_sup->srcu_barrier_completion); + wait_for_completion(&ssp->srcu_sup->srcu_barrier_completion); - rcu_seq_end(&ssp->srcu_barrier_seq); - mutex_unlock(&ssp->srcu_barrier_mutex); + rcu_seq_end(&ssp->srcu_sup->srcu_barrier_seq); + mutex_unlock(&ssp->srcu_sup->srcu_barrier_mutex); } EXPORT_SYMBOL_GPL(srcu_barrier); -- cgit From fd1b3f8e097b7fbbab8ac4a802b24fc23c703dcf Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 17 Mar 2023 21:30:32 -0700 Subject: srcu: Move work-scheduling fields from srcu_struct to srcu_usage This commit moves the ->reschedule_jiffies, ->reschedule_count, and ->work fields from the srcu_struct structure to the srcu_usage structure to reduce the size of the former in order to improve cache locality. However, this means that the container_of() calls cannot get a pointer to the srcu_struct because they are no longer in the srcu_struct. This issue is addressed by adding a ->srcu_ssp field in the srcu_usage structure that references the corresponding srcu_struct structure. And given the presence of the sup pointer to the srcu_usage structure, replace some ssp->srcu_usage-> instances with sup->. [ paulmck Apply feedback from kernel test robot. ] Link: https://lore.kernel.org/oe-kbuild-all/202303191400.iO5BOqka-lkp@intel.com/ Suggested-by: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 41 ++++++++++++++++++++++------------------- 1 file changed, 22 insertions(+), 19 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 97d1fe9a160c..169a6513b739 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -251,7 +251,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) ssp->srcu_sup->srcu_barrier_seq = 0; mutex_init(&ssp->srcu_sup->srcu_barrier_mutex); atomic_set(&ssp->srcu_sup->srcu_barrier_cpu_cnt, 0); - INIT_DELAYED_WORK(&ssp->work, process_srcu); + INIT_DELAYED_WORK(&ssp->srcu_sup->work, process_srcu); ssp->srcu_sup->sda_is_static = is_static; if (!is_static) ssp->sda = alloc_percpu(struct srcu_data); @@ -275,6 +275,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_BIG); } } + ssp->srcu_sup->srcu_ssp = ssp; smp_store_release(&ssp->srcu_sup->srcu_gp_seq_needed, 0); /* Init done. */ return 0; } @@ -647,7 +648,7 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) return; /* Just leak it! */ if (WARN_ON(srcu_readers_active(ssp))) return; /* Just leak it! */ - flush_delayed_work(&ssp->work); + flush_delayed_work(&ssp->srcu_sup->work); for_each_possible_cpu(cpu) { struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu); @@ -1059,10 +1060,10 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, // can only be executed during early boot when there is only // the one boot CPU running with interrupts still disabled. if (likely(srcu_init_done)) - queue_delayed_work(rcu_gp_wq, &ssp->work, + queue_delayed_work(rcu_gp_wq, &ssp->srcu_sup->work, !!srcu_get_delay(ssp)); - else if (list_empty(&ssp->work.work.entry)) - list_add(&ssp->work.work.entry, &srcu_boot_list); + else if (list_empty(&ssp->srcu_sup->work.work.entry)) + list_add(&ssp->srcu_sup->work.work.entry, &srcu_boot_list); } spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } @@ -1723,7 +1724,7 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) spin_unlock_irq_rcu_node(ssp->srcu_sup); if (pushgp) - queue_delayed_work(rcu_gp_wq, &ssp->work, delay); + queue_delayed_work(rcu_gp_wq, &ssp->srcu_sup->work, delay); } /* @@ -1734,22 +1735,24 @@ static void process_srcu(struct work_struct *work) unsigned long curdelay; unsigned long j; struct srcu_struct *ssp; + struct srcu_usage *sup; - ssp = container_of(work, struct srcu_struct, work.work); + sup = container_of(work, struct srcu_usage, work.work); + ssp = sup->srcu_ssp; srcu_advance_state(ssp); curdelay = srcu_get_delay(ssp); if (curdelay) { - WRITE_ONCE(ssp->reschedule_count, 0); + WRITE_ONCE(sup->reschedule_count, 0); } else { j = jiffies; - if (READ_ONCE(ssp->reschedule_jiffies) == j) { - WRITE_ONCE(ssp->reschedule_count, READ_ONCE(ssp->reschedule_count) + 1); - if (READ_ONCE(ssp->reschedule_count) > srcu_max_nodelay) + if (READ_ONCE(sup->reschedule_jiffies) == j) { + WRITE_ONCE(sup->reschedule_count, READ_ONCE(sup->reschedule_count) + 1); + if (READ_ONCE(sup->reschedule_count) > srcu_max_nodelay) curdelay = 1; } else { - WRITE_ONCE(ssp->reschedule_count, 1); - WRITE_ONCE(ssp->reschedule_jiffies, j); + WRITE_ONCE(sup->reschedule_count, 1); + WRITE_ONCE(sup->reschedule_jiffies, j); } } srcu_reschedule(ssp, curdelay); @@ -1848,7 +1851,7 @@ early_initcall(srcu_bootup_announce); void __init srcu_init(void) { - struct srcu_struct *ssp; + struct srcu_usage *sup; /* Decide on srcu_struct-size strategy. */ if (SRCU_SIZING_IS(SRCU_SIZING_AUTO)) { @@ -1868,13 +1871,13 @@ void __init srcu_init(void) */ srcu_init_done = true; while (!list_empty(&srcu_boot_list)) { - ssp = list_first_entry(&srcu_boot_list, struct srcu_struct, + sup = list_first_entry(&srcu_boot_list, struct srcu_usage, work.work.entry); - list_del_init(&ssp->work.work.entry); + list_del_init(&sup->work.work.entry); if (SRCU_SIZING_IS(SRCU_SIZING_INIT) && - ssp->srcu_sup->srcu_size_state == SRCU_SIZE_SMALL) - ssp->srcu_sup->srcu_size_state = SRCU_SIZE_ALLOC; - queue_work(rcu_gp_wq, &ssp->work.work); + sup->srcu_size_state == SRCU_SIZE_SMALL) + sup->srcu_size_state = SRCU_SIZE_ALLOC; + queue_work(rcu_gp_wq, &sup->work.work); } } -- cgit From a7bf4d7c16c1cf9753873879630a5d5169eb3206 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 24 Mar 2023 09:05:50 -0700 Subject: srcu: Check for readers at module-exit time If a given statically allocated in-module srcu_struct structure was ever used for updates, srcu_module_going() will invoke cleanup_srcu_struct() at module-exit time. This will check for the error case of SRCU readers persisting past module-exit time. On the other hand, if this srcu_struct structure never went through a grace period, srcu_module_going() only invokes free_percpu(), which would result in strange failures if SRCU readers persisted past module-exit time. This commit therefore adds a srcu_readers_active() check to srcu_module_going(), splatting if readers have persisted and refraining from invoking free_percpu() in that case. Better to leak memory than to suffer silent memory corruption! [ paulmck: Apply Zhang, Qiang1 feedback on memory leak. ] Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 169a6513b739..f9dd6ed5503e 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -1911,7 +1911,8 @@ static void srcu_module_going(struct module *mod) if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq_needed)) && !WARN_ON_ONCE(!ssp->srcu_sup->sda_is_static)) cleanup_srcu_struct(ssp); - free_percpu(ssp->sda); + if (!WARN_ON(srcu_readers_active(ssp))) + free_percpu(ssp->sda); } } -- cgit From eabe7625f053050e4cecbbe98bd944f7e8eb14f5 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 18 Mar 2023 09:34:52 -0700 Subject: srcu: Fix long lines in srcu_get_delay() This commit creates an srcu_usage pointer named "sup" as a shorter synonym for the "ssp->srcu_sup" that was bloating several lines of code. Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Cc: Christoph Hellwig Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index f9dd6ed5503e..9699bcab7215 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -616,17 +616,18 @@ static unsigned long srcu_get_delay(struct srcu_struct *ssp) unsigned long gpstart; unsigned long j; unsigned long jbase = SRCU_INTERVAL; + struct srcu_usage *sup = ssp->srcu_sup; - if (ULONG_CMP_LT(READ_ONCE(ssp->srcu_sup->srcu_gp_seq), READ_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp))) + if (ULONG_CMP_LT(READ_ONCE(sup->srcu_gp_seq), READ_ONCE(sup->srcu_gp_seq_needed_exp))) jbase = 0; - if (rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq))) { + if (rcu_seq_state(READ_ONCE(sup->srcu_gp_seq))) { j = jiffies - 1; - gpstart = READ_ONCE(ssp->srcu_sup->srcu_gp_start); + gpstart = READ_ONCE(sup->srcu_gp_start); if (time_after(j, gpstart)) jbase += j - gpstart; if (!jbase) { - WRITE_ONCE(ssp->srcu_sup->srcu_n_exp_nodelay, READ_ONCE(ssp->srcu_sup->srcu_n_exp_nodelay) + 1); - if (READ_ONCE(ssp->srcu_sup->srcu_n_exp_nodelay) > srcu_max_nodelay_phase) + WRITE_ONCE(sup->srcu_n_exp_nodelay, READ_ONCE(sup->srcu_n_exp_nodelay) + 1); + if (READ_ONCE(sup->srcu_n_exp_nodelay) > srcu_max_nodelay_phase) jbase = 1; } } -- cgit From 5ff8319f07db11c3b9347ce1dc0a6d951ae96d29 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 18 Mar 2023 10:51:50 -0700 Subject: srcu: Fix long lines in cleanup_srcu_struct() This commit creates an srcu_usage pointer named "sup" as a shorter synonym for the "ssp->srcu_sup" that was bloating several lines of code. Cc: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 9699bcab7215..11a08201ca0a 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -644,12 +644,13 @@ static unsigned long srcu_get_delay(struct srcu_struct *ssp) void cleanup_srcu_struct(struct srcu_struct *ssp) { int cpu; + struct srcu_usage *sup = ssp->srcu_sup; if (WARN_ON(!srcu_get_delay(ssp))) return; /* Just leak it! */ if (WARN_ON(srcu_readers_active(ssp))) return; /* Just leak it! */ - flush_delayed_work(&ssp->srcu_sup->work); + flush_delayed_work(&sup->work); for_each_possible_cpu(cpu) { struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu); @@ -658,21 +659,21 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) if (WARN_ON(rcu_segcblist_n_cbs(&sdp->srcu_cblist))) return; /* Forgot srcu_barrier(), so just leak it! */ } - if (WARN_ON(rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)) != SRCU_STATE_IDLE) || - WARN_ON(rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq) != ssp->srcu_sup->srcu_gp_seq_needed) || + if (WARN_ON(rcu_seq_state(READ_ONCE(sup->srcu_gp_seq)) != SRCU_STATE_IDLE) || + WARN_ON(rcu_seq_current(&sup->srcu_gp_seq) != sup->srcu_gp_seq_needed) || WARN_ON(srcu_readers_active(ssp))) { pr_info("%s: Active srcu_struct %p read state: %d gp state: %lu/%lu\n", - __func__, ssp, rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)), - rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq), ssp->srcu_sup->srcu_gp_seq_needed); + __func__, ssp, rcu_seq_state(READ_ONCE(sup->srcu_gp_seq)), + rcu_seq_current(&sup->srcu_gp_seq), sup->srcu_gp_seq_needed); return; /* Caller forgot to stop doing call_srcu()? */ } - kfree(ssp->srcu_sup->node); - ssp->srcu_sup->node = NULL; - ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; - if (!ssp->srcu_sup->sda_is_static) { + kfree(sup->node); + sup->node = NULL; + sup->srcu_size_state = SRCU_SIZE_SMALL; + if (!sup->sda_is_static) { free_percpu(ssp->sda); ssp->sda = NULL; - kfree(ssp->srcu_sup); + kfree(sup); ssp->srcu_sup = NULL; } } -- cgit From 6c366522e10f2b5e916b3f08ac042df1a1cd512a Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 18 Mar 2023 10:52:48 -0700 Subject: srcu: Fix long lines in srcu_gp_end() This commit creates an srcu_usage pointer named "sup" as a shorter synonym for the "ssp->srcu_sup" that was bloating several lines of code. Cc: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 41 +++++++++++++++++++++-------------------- 1 file changed, 21 insertions(+), 20 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 11a08201ca0a..f661a0f6bc0d 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -862,28 +862,29 @@ static void srcu_gp_end(struct srcu_struct *ssp) unsigned long sgsne; struct srcu_node *snp; int ss_state; + struct srcu_usage *sup = ssp->srcu_sup; /* Prevent more than one additional grace period. */ - mutex_lock(&ssp->srcu_sup->srcu_cb_mutex); + mutex_lock(&sup->srcu_cb_mutex); /* End the current grace period. */ - spin_lock_irq_rcu_node(ssp->srcu_sup); - idx = rcu_seq_state(ssp->srcu_sup->srcu_gp_seq); + spin_lock_irq_rcu_node(sup); + idx = rcu_seq_state(sup->srcu_gp_seq); WARN_ON_ONCE(idx != SRCU_STATE_SCAN2); - if (ULONG_CMP_LT(READ_ONCE(ssp->srcu_sup->srcu_gp_seq), READ_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp))) + if (ULONG_CMP_LT(READ_ONCE(sup->srcu_gp_seq), READ_ONCE(sup->srcu_gp_seq_needed_exp))) cbdelay = 0; - WRITE_ONCE(ssp->srcu_sup->srcu_last_gp_end, ktime_get_mono_fast_ns()); - rcu_seq_end(&ssp->srcu_sup->srcu_gp_seq); - gpseq = rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq); - if (ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed_exp, gpseq)) - WRITE_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp, gpseq); - spin_unlock_irq_rcu_node(ssp->srcu_sup); - mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); + WRITE_ONCE(sup->srcu_last_gp_end, ktime_get_mono_fast_ns()); + rcu_seq_end(&sup->srcu_gp_seq); + gpseq = rcu_seq_current(&sup->srcu_gp_seq); + if (ULONG_CMP_LT(sup->srcu_gp_seq_needed_exp, gpseq)) + WRITE_ONCE(sup->srcu_gp_seq_needed_exp, gpseq); + spin_unlock_irq_rcu_node(sup); + mutex_unlock(&sup->srcu_gp_mutex); /* A new grace period can start at this point. But only one. */ /* Initiate callback invocation as needed. */ - ss_state = smp_load_acquire(&ssp->srcu_sup->srcu_size_state); + ss_state = smp_load_acquire(&sup->srcu_size_state); if (ss_state < SRCU_SIZE_WAIT_BARRIER) { srcu_schedule_cbs_sdp(per_cpu_ptr(ssp->sda, get_boot_cpu_id()), cbdelay); @@ -892,7 +893,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) srcu_for_each_node_breadth_first(ssp, snp) { spin_lock_irq_rcu_node(snp); cbs = false; - last_lvl = snp >= ssp->srcu_sup->level[rcu_num_lvls - 1]; + last_lvl = snp >= sup->level[rcu_num_lvls - 1]; if (last_lvl) cbs = ss_state < SRCU_SIZE_BIG || snp->srcu_have_cbs[idx] == gpseq; snp->srcu_have_cbs[idx] = gpseq; @@ -924,18 +925,18 @@ static void srcu_gp_end(struct srcu_struct *ssp) } /* Callback initiation done, allow grace periods after next. */ - mutex_unlock(&ssp->srcu_sup->srcu_cb_mutex); + mutex_unlock(&sup->srcu_cb_mutex); /* Start a new grace period if needed. */ - spin_lock_irq_rcu_node(ssp->srcu_sup); - gpseq = rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq); + spin_lock_irq_rcu_node(sup); + gpseq = rcu_seq_current(&sup->srcu_gp_seq); if (!rcu_seq_state(gpseq) && - ULONG_CMP_LT(gpseq, ssp->srcu_sup->srcu_gp_seq_needed)) { + ULONG_CMP_LT(gpseq, sup->srcu_gp_seq_needed)) { srcu_gp_start(ssp); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + spin_unlock_irq_rcu_node(sup); srcu_reschedule(ssp, 0); } else { - spin_unlock_irq_rcu_node(ssp->srcu_sup); + spin_unlock_irq_rcu_node(sup); } /* Transition to big if needed. */ @@ -943,7 +944,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) if (ss_state == SRCU_SIZE_ALLOC) init_srcu_struct_nodes(ssp, GFP_KERNEL); else - smp_store_release(&ssp->srcu_sup->srcu_size_state, ss_state + 1); + smp_store_release(&sup->srcu_size_state, ss_state + 1); } } -- cgit From cefc0a599b19d8dd0e26d0b2e43311bae7530ca1 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 18 Mar 2023 12:31:53 -0700 Subject: srcu: Fix long lines in srcu_funnel_gp_start() This commit creates an srcu_usage pointer named "sup" as a shorter synonym for the "ssp->srcu_sup" that was bloating several lines of code. Cc: Christoph Hellwig Tested-by: Sachin Sant Tested-by: "Zhang, Qiang1" Tested-by: Joel Fernandes (Google) Signed-off-by: Paul E. McKenney --- kernel/rcu/srcutree.c | 27 ++++++++++++++------------- 1 file changed, 14 insertions(+), 13 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index f661a0f6bc0d..a887cfc89894 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -1004,9 +1004,10 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, struct srcu_node *snp; struct srcu_node *snp_leaf; unsigned long snp_seq; + struct srcu_usage *sup = ssp->srcu_sup; /* Ensure that snp node tree is fully initialized before traversing it */ - if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) + if (smp_load_acquire(&sup->srcu_size_state) < SRCU_SIZE_WAIT_BARRIER) snp_leaf = NULL; else snp_leaf = sdp->mynode; @@ -1014,7 +1015,7 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, if (snp_leaf) /* Each pass through the loop does one level of the srcu_node tree. */ for (snp = snp_leaf; snp != NULL; snp = snp->srcu_parent) { - if (WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, s)) && snp != snp_leaf) + if (WARN_ON_ONCE(rcu_seq_done(&sup->srcu_gp_seq, s)) && snp != snp_leaf) return; /* GP already done and CBs recorded. */ spin_lock_irqsave_rcu_node(snp, flags); snp_seq = snp->srcu_have_cbs[idx]; @@ -1041,20 +1042,20 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, /* Top of tree, must ensure the grace period will be started. */ spin_lock_irqsave_ssp_contention(ssp, &flags); - if (ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed, s)) { + if (ULONG_CMP_LT(sup->srcu_gp_seq_needed, s)) { /* * Record need for grace period s. Pair with load * acquire setting up for initialization. */ - smp_store_release(&ssp->srcu_sup->srcu_gp_seq_needed, s); /*^^^*/ + smp_store_release(&sup->srcu_gp_seq_needed, s); /*^^^*/ } - if (!do_norm && ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed_exp, s)) - WRITE_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp, s); + if (!do_norm && ULONG_CMP_LT(sup->srcu_gp_seq_needed_exp, s)) + WRITE_ONCE(sup->srcu_gp_seq_needed_exp, s); /* If grace period not already in progress, start it. */ - if (!WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, s)) && - rcu_seq_state(ssp->srcu_sup->srcu_gp_seq) == SRCU_STATE_IDLE) { - WARN_ON_ONCE(ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)); + if (!WARN_ON_ONCE(rcu_seq_done(&sup->srcu_gp_seq, s)) && + rcu_seq_state(sup->srcu_gp_seq) == SRCU_STATE_IDLE) { + WARN_ON_ONCE(ULONG_CMP_GE(sup->srcu_gp_seq, sup->srcu_gp_seq_needed)); srcu_gp_start(ssp); // And how can that list_add() in the "else" clause @@ -1063,12 +1064,12 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, // can only be executed during early boot when there is only // the one boot CPU running with interrupts still disabled. if (likely(srcu_init_done)) - queue_delayed_work(rcu_gp_wq, &ssp->srcu_sup->work, + queue_delayed_work(rcu_gp_wq, &sup->work, !!srcu_get_delay(ssp)); - else if (list_empty(&ssp->srcu_sup->work.work.entry)) - list_add(&ssp->srcu_sup->work.work.entry, &srcu_boot_list); + else if (list_empty(&sup->work.work.entry)) + list_add(&sup->work.work.entry, &srcu_boot_list); } - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + spin_unlock_irqrestore_rcu_node(sup, flags); } /* -- cgit From 754aa6427efeb8a059233e18e810263a108fdd71 Mon Sep 17 00:00:00 2001 From: "Joel Fernandes (Google)" Date: Sat, 28 Jan 2023 03:59:01 +0000 Subject: srcu: Clarify comments on memory barrier "E" There is an smp_mb() named "E" in srcu_flip() immediately before the increment (flip) of the srcu_struct structure's ->srcu_idx. The purpose of E is to order the preceding scan's read of lock counters against the flipping of the ->srcu_idx, in order to prevent new readers from continuing to use the old ->srcu_idx value, which might needlessly extend the grace period. However, this ordering is already enforced because of the control dependency between the preceding scan and the ->srcu_idx flip. This control dependency exists because atomic_long_read() is used to scan the counts, because WRITE_ONCE() is used to flip ->srcu_idx, and because ->srcu_idx is not flipped until the ->srcu_lock_count[] and ->srcu_unlock_count[] counts match. And such a match cannot happen when there is an in-flight reader that started before the flip (observation courtesy Mathieu Desnoyers). The litmus test below (courtesy of Frederic Weisbecker, with changes for ctrldep by Boqun and Joel) shows this: C srcu (* * bad condition: P0's first scan (SCAN1) saw P1's idx=0 LOCK count inc, though P1 saw flip. * * So basically, the ->po ordering on both P0 and P1 is enforced via ->ppo * (control deps) on both sides, and both P0 and P1 are interconnected by ->rf * relations. Combining the ->ppo with ->rf, a cycle is impossible. *) {} // updater P0(int *IDX, int *LOCK0, int *UNLOCK0, int *LOCK1, int *UNLOCK1) { int lock1; int unlock1; int lock0; int unlock0; // SCAN1 unlock1 = READ_ONCE(*UNLOCK1); smp_mb(); // A lock1 = READ_ONCE(*LOCK1); // FLIP if (lock1 == unlock1) { // Control dep smp_mb(); // E // Remove E and still passes. WRITE_ONCE(*IDX, 1); smp_mb(); // D // SCAN2 unlock0 = READ_ONCE(*UNLOCK0); smp_mb(); // A lock0 = READ_ONCE(*LOCK0); } } // reader P1(int *IDX, int *LOCK0, int *UNLOCK0, int *LOCK1, int *UNLOCK1) { int tmp; int idx1; int idx2; // 1st reader idx1 = READ_ONCE(*IDX); if (idx1 == 0) { // Control dep tmp = READ_ONCE(*LOCK0); WRITE_ONCE(*LOCK0, tmp + 1); smp_mb(); /* B and C */ tmp = READ_ONCE(*UNLOCK0); WRITE_ONCE(*UNLOCK0, tmp + 1); } else { tmp = READ_ONCE(*LOCK1); WRITE_ONCE(*LOCK1, tmp + 1); smp_mb(); /* B and C */ tmp = READ_ONCE(*UNLOCK1); WRITE_ONCE(*UNLOCK1, tmp + 1); } } exists (0:lock1=1 /\ 1:idx1=1) More complicated litmus tests with multiple SRCU readers also show that memory barrier E is not needed. This commit therefore clarifies the comment on memory barrier E. Why not also remove that redundant smp_mb()? Because control dependencies are quite fragile due to their not being recognized by most compilers and tools. Control dependencies therefore exact an ongoing maintenance burden, and such a burden cannot be justified in this slowpath. Therefore, that smp_mb() stays until such time as its overhead becomes a measurable problem in a real workload running on a real production system, or until such time as compilers start paying attention to this sort of control dependency. Co-developed-by: Frederic Weisbecker Signed-off-by: Frederic Weisbecker Co-developed-by: Mathieu Desnoyers Signed-off-by: Mathieu Desnoyers Co-developed-by: Boqun Feng Signed-off-by: Boqun Feng Reviewed-by: Paul E. McKenney Signed-off-by: Joel Fernandes (Google) --- kernel/rcu/srcutree.c | 34 +++++++++++++++++++++++++++------- 1 file changed, 27 insertions(+), 7 deletions(-) (limited to 'kernel/rcu/srcutree.c') diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index ab4ee58af84b..68f89c533057 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -1085,16 +1085,36 @@ static bool try_check_zero(struct srcu_struct *ssp, int idx, int trycount) static void srcu_flip(struct srcu_struct *ssp) { /* - * Ensure that if this updater saw a given reader's increment - * from __srcu_read_lock(), that reader was using an old value - * of ->srcu_idx. Also ensure that if a given reader sees the - * new value of ->srcu_idx, this updater's earlier scans cannot - * have seen that reader's increments (which is OK, because this - * grace period need not wait on that reader). + * Because the flip of ->srcu_idx is executed only if the + * preceding call to srcu_readers_active_idx_check() found that + * the ->srcu_unlock_count[] and ->srcu_lock_count[] sums matched + * and because that summing uses atomic_long_read(), there is + * ordering due to a control dependency between that summing and + * the WRITE_ONCE() in this call to srcu_flip(). This ordering + * ensures that if this updater saw a given reader's increment from + * __srcu_read_lock(), that reader was using a value of ->srcu_idx + * from before the previous call to srcu_flip(), which should be + * quite rare. This ordering thus helps forward progress because + * the grace period could otherwise be delayed by additional + * calls to __srcu_read_lock() using that old (soon to be new) + * value of ->srcu_idx. + * + * This sum-equality check and ordering also ensures that if + * a given call to __srcu_read_lock() uses the new value of + * ->srcu_idx, this updater's earlier scans cannot have seen + * that reader's increments, which is all to the good, because + * this grace period need not wait on that reader. After all, + * if those earlier scans had seen that reader, there would have + * been a sum mismatch and this code would not be reached. + * + * This means that the following smp_mb() is redundant, but + * it stays until either (1) Compilers learn about this sort of + * control dependency or (2) Some production workload running on + * a production system is unduly delayed by this slowpath smp_mb(). */ smp_mb(); /* E */ /* Pairs with B and C. */ - WRITE_ONCE(ssp->srcu_idx, ssp->srcu_idx + 1); + WRITE_ONCE(ssp->srcu_idx, ssp->srcu_idx + 1); // Flip the counter. /* * Ensure that if the updater misses an __srcu_read_unlock() -- cgit