From 95846ecf9dac5089aed4b144d912225f8ef86ae4 Mon Sep 17 00:00:00 2001 From: Gargi Sharma Date: Fri, 17 Nov 2017 15:30:30 -0800 Subject: pid: replace pid bitmap implementation with IDR API Patch series "Replacing PID bitmap implementation with IDR API", v4. This series replaces kernel bitmap implementation of PID allocation with IDR API. These patches are written to simplify the kernel by replacing custom code with calls to generic code. The following are the stats for pid and pid_namespace object files before and after the replacement. There is a noteworthy change between the IDR and bitmap implementation. Before text data bss dec hex filename 8447 3894 64 12405 3075 kernel/pid.o After text data bss dec hex filename 3397 304 0 3701 e75 kernel/pid.o Before text data bss dec hex filename 5692 1842 192 7726 1e2e kernel/pid_namespace.o After text data bss dec hex filename 2854 216 16 3086 c0e kernel/pid_namespace.o The following are the stats for ps, pstree and calling readdir on /proc for 10,000 processes. ps: With IDR API With bitmap real 0m1.479s 0m2.319s user 0m0.070s 0m0.060s sys 0m0.289s 0m0.516s pstree: With IDR API With bitmap real 0m1.024s 0m1.794s user 0m0.348s 0m0.612s sys 0m0.184s 0m0.264s proc: With IDR API With bitmap real 0m0.059s 0m0.074s user 0m0.000s 0m0.004s sys 0m0.016s 0m0.016s This patch (of 2): Replace the current bitmap implementation for Process ID allocation. Functions that are no longer required, for example, free_pidmap(), alloc_pidmap(), etc. are removed. The rest of the functions are modified to use the IDR API. The change was made to make the PID allocation less complex by replacing custom code with calls to generic API. [gs051095@gmail.com: v6] Link: http://lkml.kernel.org/r/1507760379-21662-2-git-send-email-gs051095@gmail.com [avagin@openvz.org: restore the old behaviour of the ns_last_pid sysctl] Link: http://lkml.kernel.org/r/20171106183144.16368-1-avagin@openvz.org Link: http://lkml.kernel.org/r/1507583624-22146-2-git-send-email-gs051095@gmail.com Signed-off-by: Gargi Sharma Reviewed-by: Rik van Riel Acked-by: Oleg Nesterov Cc: Julia Lawall Cc: Ingo Molnar Cc: Pavel Tatashin Cc: Kirill Tkhai Cc: Eric W. Biederman Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/pid.c | 201 +++++++++++------------------------------------------------ 1 file changed, 36 insertions(+), 165 deletions(-) (limited to 'kernel/pid.c') diff --git a/kernel/pid.c b/kernel/pid.c index 020dedbdf066..0ce59369632f 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -39,6 +39,7 @@ #include #include #include +#include #define pid_hashfn(nr, ns) \ hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift) @@ -53,14 +54,6 @@ int pid_max = PID_MAX_DEFAULT; int pid_max_min = RESERVED_PIDS + 1; int pid_max_max = PID_MAX_LIMIT; -static inline int mk_pid(struct pid_namespace *pid_ns, - struct pidmap *map, int off) -{ - return (map - pid_ns->pidmap)*BITS_PER_PAGE + off; -} - -#define find_next_offset(map, off) \ - find_next_zero_bit((map)->page, BITS_PER_PAGE, off) /* * PID-map pages start out as NULL, they get allocated upon @@ -70,10 +63,7 @@ static inline int mk_pid(struct pid_namespace *pid_ns, */ struct pid_namespace init_pid_ns = { .kref = KREF_INIT(2), - .pidmap = { - [ 0 ... PIDMAP_ENTRIES-1] = { ATOMIC_INIT(BITS_PER_PAGE), NULL } - }, - .last_pid = 0, + .idr = IDR_INIT, .nr_hashed = PIDNS_HASH_ADDING, .level = 0, .child_reaper = &init_task, @@ -101,138 +91,6 @@ EXPORT_SYMBOL_GPL(init_pid_ns); static __cacheline_aligned_in_smp DEFINE_SPINLOCK(pidmap_lock); -static void free_pidmap(struct upid *upid) -{ - int nr = upid->nr; - struct pidmap *map = upid->ns->pidmap + nr / BITS_PER_PAGE; - int offset = nr & BITS_PER_PAGE_MASK; - - clear_bit(offset, map->page); - atomic_inc(&map->nr_free); -} - -/* - * If we started walking pids at 'base', is 'a' seen before 'b'? - */ -static int pid_before(int base, int a, int b) -{ - /* - * This is the same as saying - * - * (a - base + MAXUINT) % MAXUINT < (b - base + MAXUINT) % MAXUINT - * and that mapping orders 'a' and 'b' with respect to 'base'. - */ - return (unsigned)(a - base) < (unsigned)(b - base); -} - -/* - * We might be racing with someone else trying to set pid_ns->last_pid - * at the pid allocation time (there's also a sysctl for this, but racing - * with this one is OK, see comment in kernel/pid_namespace.c about it). - * We want the winner to have the "later" value, because if the - * "earlier" value prevails, then a pid may get reused immediately. - * - * Since pids rollover, it is not sufficient to just pick the bigger - * value. We have to consider where we started counting from. - * - * 'base' is the value of pid_ns->last_pid that we observed when - * we started looking for a pid. - * - * 'pid' is the pid that we eventually found. - */ -static void set_last_pid(struct pid_namespace *pid_ns, int base, int pid) -{ - int prev; - int last_write = base; - do { - prev = last_write; - last_write = cmpxchg(&pid_ns->last_pid, prev, pid); - } while ((prev != last_write) && (pid_before(base, last_write, pid))); -} - -static int alloc_pidmap(struct pid_namespace *pid_ns) -{ - int i, offset, max_scan, pid, last = pid_ns->last_pid; - struct pidmap *map; - - pid = last + 1; - if (pid >= pid_max) - pid = RESERVED_PIDS; - offset = pid & BITS_PER_PAGE_MASK; - map = &pid_ns->pidmap[pid/BITS_PER_PAGE]; - /* - * If last_pid points into the middle of the map->page we - * want to scan this bitmap block twice, the second time - * we start with offset == 0 (or RESERVED_PIDS). - */ - max_scan = DIV_ROUND_UP(pid_max, BITS_PER_PAGE) - !offset; - for (i = 0; i <= max_scan; ++i) { - if (unlikely(!map->page)) { - void *page = kzalloc(PAGE_SIZE, GFP_KERNEL); - /* - * Free the page if someone raced with us - * installing it: - */ - spin_lock_irq(&pidmap_lock); - if (!map->page) { - map->page = page; - page = NULL; - } - spin_unlock_irq(&pidmap_lock); - kfree(page); - if (unlikely(!map->page)) - return -ENOMEM; - } - if (likely(atomic_read(&map->nr_free))) { - for ( ; ; ) { - if (!test_and_set_bit(offset, map->page)) { - atomic_dec(&map->nr_free); - set_last_pid(pid_ns, last, pid); - return pid; - } - offset = find_next_offset(map, offset); - if (offset >= BITS_PER_PAGE) - break; - pid = mk_pid(pid_ns, map, offset); - if (pid >= pid_max) - break; - } - } - if (map < &pid_ns->pidmap[(pid_max-1)/BITS_PER_PAGE]) { - ++map; - offset = 0; - } else { - map = &pid_ns->pidmap[0]; - offset = RESERVED_PIDS; - if (unlikely(last == offset)) - break; - } - pid = mk_pid(pid_ns, map, offset); - } - return -EAGAIN; -} - -int next_pidmap(struct pid_namespace *pid_ns, unsigned int last) -{ - int offset; - struct pidmap *map, *end; - - if (last >= PID_MAX_LIMIT) - return -1; - - offset = (last + 1) & BITS_PER_PAGE_MASK; - map = &pid_ns->pidmap[(last + 1)/BITS_PER_PAGE]; - end = &pid_ns->pidmap[PIDMAP_ENTRIES]; - for (; map < end; map++, offset = 0) { - if (unlikely(!map->page)) - continue; - offset = find_next_bit((map)->page, BITS_PER_PAGE, offset); - if (offset < BITS_PER_PAGE) - return mk_pid(pid_ns, map, offset); - } - return -1; -} - void put_pid(struct pid *pid) { struct pid_namespace *ns; @@ -266,7 +124,7 @@ void free_pid(struct pid *pid) struct upid *upid = pid->numbers + i; struct pid_namespace *ns = upid->ns; hlist_del_rcu(&upid->pid_chain); - switch(--ns->nr_hashed) { + switch (--ns->nr_hashed) { case 2: case 1: /* When all that is left in the pid namespace @@ -284,12 +142,11 @@ void free_pid(struct pid *pid) schedule_work(&ns->proc_work); break; } + + idr_remove(&ns->idr, upid->nr); } spin_unlock_irqrestore(&pidmap_lock, flags); - for (i = 0; i <= pid->level; i++) - free_pidmap(pid->numbers + i); - call_rcu(&pid->rcu, delayed_put_pid); } @@ -308,8 +165,29 @@ struct pid *alloc_pid(struct pid_namespace *ns) tmp = ns; pid->level = ns->level; + for (i = ns->level; i >= 0; i--) { - nr = alloc_pidmap(tmp); + int pid_min = 1; + + idr_preload(GFP_KERNEL); + spin_lock_irq(&pidmap_lock); + + /* + * init really needs pid 1, but after reaching the maximum + * wrap back to RESERVED_PIDS + */ + if (idr_get_cursor(&tmp->idr) > RESERVED_PIDS) + pid_min = RESERVED_PIDS; + + /* + * Store a null pointer so find_pid_ns does not find + * a partially initialized PID (see below). + */ + nr = idr_alloc_cyclic(&tmp->idr, NULL, pid_min, + pid_max, GFP_ATOMIC); + spin_unlock_irq(&pidmap_lock); + idr_preload_end(); + if (nr < 0) { retval = nr; goto out_free; @@ -339,6 +217,8 @@ struct pid *alloc_pid(struct pid_namespace *ns) for ( ; upid >= pid->numbers; --upid) { hlist_add_head_rcu(&upid->pid_chain, &pid_hash[pid_hashfn(upid->nr, upid->ns)]); + /* Make the PID visible to find_pid_ns. */ + idr_replace(&upid->ns->idr, pid, upid->nr); upid->ns->nr_hashed++; } spin_unlock_irq(&pidmap_lock); @@ -350,8 +230,11 @@ out_unlock: put_pid_ns(ns); out_free: + spin_lock_irq(&pidmap_lock); while (++i <= ns->level) - free_pidmap(pid->numbers + i); + idr_remove(&ns->idr, (pid->numbers + i)->nr); + + spin_unlock_irq(&pidmap_lock); kmem_cache_free(ns->pid_cachep, pid); return ERR_PTR(retval); @@ -553,16 +436,7 @@ EXPORT_SYMBOL_GPL(task_active_pid_ns); */ struct pid *find_ge_pid(int nr, struct pid_namespace *ns) { - struct pid *pid; - - do { - pid = find_pid_ns(nr, ns); - if (pid) - break; - nr = next_pidmap(ns, nr); - } while (nr > 0); - - return pid; + return idr_get_next(&ns->idr, &nr); } /* @@ -578,7 +452,7 @@ void __init pidhash_init(void) 0, 4096); } -void __init pidmap_init(void) +void __init pid_idr_init(void) { /* Verify no one has done anything silly: */ BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_HASH_ADDING); @@ -590,10 +464,7 @@ void __init pidmap_init(void) PIDS_PER_CPU_MIN * num_possible_cpus()); pr_info("pid_max: default: %u minimum: %u\n", pid_max, pid_max_min); - init_pid_ns.pidmap[0].page = kzalloc(PAGE_SIZE, GFP_KERNEL); - /* Reserve PID 0. We never call free_pidmap(0) */ - set_bit(0, init_pid_ns.pidmap[0].page); - atomic_dec(&init_pid_ns.pidmap[0].nr_free); + idr_init(&init_pid_ns.idr); init_pid_ns.pid_cachep = KMEM_CACHE(pid, SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT); -- cgit v1.3-8-gc7d7 From e8cfbc245e24887e3c30235f71e9e9405e0cfc39 Mon Sep 17 00:00:00 2001 From: Gargi Sharma Date: Fri, 17 Nov 2017 15:30:34 -0800 Subject: pid: remove pidhash pidhash is no longer required as all the information can be looked up from idr tree. nr_hashed represented the number of pids that had been hashed. Since, nr_hashed and PIDNS_HASH_ADDING are no longer relevant, it has been renamed to pid_allocated and PIDNS_ADDING respectively. [gs051095@gmail.com: v6] Link: http://lkml.kernel.org/r/1507760379-21662-3-git-send-email-gs051095@gmail.com Link: http://lkml.kernel.org/r/1507583624-22146-3-git-send-email-gs051095@gmail.com Signed-off-by: Gargi Sharma Reviewed-by: Rik van Riel Tested-by: Tony Luck [ia64] Cc: Julia Lawall Cc: Ingo Molnar Cc: Pavel Tatashin Cc: Kirill Tkhai Cc: Oleg Nesterov Cc: Eric W. Biederman Cc: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/asm-offsets.c | 4 ++-- include/linux/init_task.h | 1 - include/linux/pid.h | 2 -- include/linux/pid_namespace.h | 4 ++-- init/main.c | 1 - kernel/fork.c | 2 +- kernel/pid.c | 48 +++++++++--------------------------------- kernel/pid_namespace.c | 6 +++--- 8 files changed, 18 insertions(+), 50 deletions(-) (limited to 'kernel/pid.c') diff --git a/arch/ia64/kernel/asm-offsets.c b/arch/ia64/kernel/asm-offsets.c index f7693f49c573..f4db2168d1b8 100644 --- a/arch/ia64/kernel/asm-offsets.c +++ b/arch/ia64/kernel/asm-offsets.c @@ -31,8 +31,8 @@ void foo(void) DEFINE(SIGFRAME_SIZE, sizeof (struct sigframe)); DEFINE(UNW_FRAME_INFO_SIZE, sizeof (struct unw_frame_info)); - BUILD_BUG_ON(sizeof(struct upid) != 32); - DEFINE(IA64_UPID_SHIFT, 5); + BUILD_BUG_ON(sizeof(struct upid) != 16); + DEFINE(IA64_UPID_SHIFT, 4); BLANK(); diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 8062e6cc607c..6a532629c983 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -105,7 +105,6 @@ extern struct group_info init_groups; .numbers = { { \ .nr = 0, \ .ns = &init_pid_ns, \ - .pid_chain = { .next = NULL, .pprev = NULL }, \ }, } \ } diff --git a/include/linux/pid.h b/include/linux/pid.h index dfd684ce0787..7633d55d9a24 100644 --- a/include/linux/pid.h +++ b/include/linux/pid.h @@ -51,10 +51,8 @@ enum pid_type */ struct upid { - /* Try to keep pid_chain in the same cacheline as nr for find_vpid */ int nr; struct pid_namespace *ns; - struct hlist_node pid_chain; }; struct pid diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h index 92c6aa509d2e..49538b172483 100644 --- a/include/linux/pid_namespace.h +++ b/include/linux/pid_namespace.h @@ -25,7 +25,7 @@ struct pid_namespace { struct kref kref; struct idr idr; struct rcu_head rcu; - unsigned int nr_hashed; + unsigned int pid_allocated; struct task_struct *child_reaper; struct kmem_cache *pid_cachep; unsigned int level; @@ -49,7 +49,7 @@ struct pid_namespace { extern struct pid_namespace init_pid_ns; -#define PIDNS_HASH_ADDING (1U << 31) +#define PIDNS_ADDING (1U << 31) #ifdef CONFIG_PID_NS static inline struct pid_namespace *get_pid_ns(struct pid_namespace *ns) diff --git a/init/main.c b/init/main.c index d0cbcfc06124..dfec3809e740 100644 --- a/init/main.c +++ b/init/main.c @@ -562,7 +562,6 @@ asmlinkage __visible void __init start_kernel(void) * kmem_cache_init() */ setup_log_buf(0); - pidhash_init(); vfs_caches_init_early(); sort_main_extable(); trap_init(); diff --git a/kernel/fork.c b/kernel/fork.c index 4e55eedba8d6..432eadf6b58c 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1871,7 +1871,7 @@ static __latent_entropy struct task_struct *copy_process( retval = -ERESTARTNOINTR; goto bad_fork_cancel_cgroup; } - if (unlikely(!(ns_of_pid(pid)->nr_hashed & PIDNS_HASH_ADDING))) { + if (unlikely(!(ns_of_pid(pid)->pid_allocated & PIDNS_ADDING))) { retval = -ENOMEM; goto bad_fork_cancel_cgroup; } diff --git a/kernel/pid.c b/kernel/pid.c index 0ce59369632f..b13b624e2c49 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -41,10 +41,6 @@ #include #include -#define pid_hashfn(nr, ns) \ - hash_long((unsigned long)nr + (unsigned long)ns, pidhash_shift) -static struct hlist_head *pid_hash; -static unsigned int pidhash_shift = 4; struct pid init_struct_pid = INIT_STRUCT_PID; int pid_max = PID_MAX_DEFAULT; @@ -54,7 +50,6 @@ int pid_max = PID_MAX_DEFAULT; int pid_max_min = RESERVED_PIDS + 1; int pid_max_max = PID_MAX_LIMIT; - /* * PID-map pages start out as NULL, they get allocated upon * first use and are never deallocated. This way a low pid_max @@ -64,7 +59,7 @@ int pid_max_max = PID_MAX_LIMIT; struct pid_namespace init_pid_ns = { .kref = KREF_INIT(2), .idr = IDR_INIT, - .nr_hashed = PIDNS_HASH_ADDING, + .pid_allocated = PIDNS_ADDING, .level = 0, .child_reaper = &init_task, .user_ns = &init_user_ns, @@ -123,8 +118,7 @@ void free_pid(struct pid *pid) for (i = 0; i <= pid->level; i++) { struct upid *upid = pid->numbers + i; struct pid_namespace *ns = upid->ns; - hlist_del_rcu(&upid->pid_chain); - switch (--ns->nr_hashed) { + switch (--ns->pid_allocated) { case 2: case 1: /* When all that is left in the pid namespace @@ -133,10 +127,10 @@ void free_pid(struct pid *pid) */ wake_up_process(ns->child_reaper); break; - case PIDNS_HASH_ADDING: + case PIDNS_ADDING: /* Handle a fork failure of the first process */ WARN_ON(ns->child_reaper); - ns->nr_hashed = 0; + ns->pid_allocated = 0; /* fall through */ case 0: schedule_work(&ns->proc_work); @@ -212,14 +206,12 @@ struct pid *alloc_pid(struct pid_namespace *ns) upid = pid->numbers + ns->level; spin_lock_irq(&pidmap_lock); - if (!(ns->nr_hashed & PIDNS_HASH_ADDING)) + if (!(ns->pid_allocated & PIDNS_ADDING)) goto out_unlock; for ( ; upid >= pid->numbers; --upid) { - hlist_add_head_rcu(&upid->pid_chain, - &pid_hash[pid_hashfn(upid->nr, upid->ns)]); /* Make the PID visible to find_pid_ns. */ idr_replace(&upid->ns->idr, pid, upid->nr); - upid->ns->nr_hashed++; + upid->ns->pid_allocated++; } spin_unlock_irq(&pidmap_lock); @@ -243,21 +235,13 @@ out_free: void disable_pid_allocation(struct pid_namespace *ns) { spin_lock_irq(&pidmap_lock); - ns->nr_hashed &= ~PIDNS_HASH_ADDING; + ns->pid_allocated &= ~PIDNS_ADDING; spin_unlock_irq(&pidmap_lock); } struct pid *find_pid_ns(int nr, struct pid_namespace *ns) { - struct upid *pnr; - - hlist_for_each_entry_rcu(pnr, - &pid_hash[pid_hashfn(nr, ns)], pid_chain) - if (pnr->nr == nr && pnr->ns == ns) - return container_of(pnr, struct pid, - numbers[ns->level]); - - return NULL; + return idr_find(&ns->idr, nr); } EXPORT_SYMBOL_GPL(find_pid_ns); @@ -413,6 +397,7 @@ pid_t __task_pid_nr_ns(struct task_struct *task, enum pid_type type, if (type != PIDTYPE_PID) { if (type == __PIDTYPE_TGID) type = PIDTYPE_PID; + task = task->group_leader; } nr = pid_nr_ns(rcu_dereference(task->pids[type].pid), ns); @@ -439,23 +424,10 @@ struct pid *find_ge_pid(int nr, struct pid_namespace *ns) return idr_get_next(&ns->idr, &nr); } -/* - * The pid hash table is scaled according to the amount of memory in the - * machine. From a minimum of 16 slots up to 4096 slots at one gigabyte or - * more. - */ -void __init pidhash_init(void) -{ - pid_hash = alloc_large_system_hash("PID", sizeof(*pid_hash), 0, 18, - HASH_EARLY | HASH_SMALL | HASH_ZERO, - &pidhash_shift, NULL, - 0, 4096); -} - void __init pid_idr_init(void) { /* Verify no one has done anything silly: */ - BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_HASH_ADDING); + BUILD_BUG_ON(PID_MAX_LIMIT >= PIDNS_ADDING); /* bump default and minimum pid_max based on number of cpus */ pid_max = min(pid_max_max, max_t(int, pid_max, diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c index ca7c8a8823b1..0b53eef7d34b 100644 --- a/kernel/pid_namespace.c +++ b/kernel/pid_namespace.c @@ -133,7 +133,7 @@ static struct pid_namespace *create_pid_namespace(struct user_namespace *user_ns ns->parent = get_pid_ns(parent_pid_ns); ns->user_ns = get_user_ns(user_ns); ns->ucounts = ucounts; - ns->nr_hashed = PIDNS_HASH_ADDING; + ns->pid_allocated = PIDNS_ADDING; INIT_WORK(&ns->proc_work, proc_cleanup_work); return ns; @@ -254,7 +254,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns) * sys_wait4() above can't reap the EXIT_DEAD children but we do not * really care, we could reparent them to the global init. We could * exit and reap ->child_reaper even if it is not the last thread in - * this pid_ns, free_pid(nr_hashed == 0) calls proc_cleanup_work(), + * this pid_ns, free_pid(pid_allocated == 0) calls proc_cleanup_work(), * pid_ns can not go away until proc_kill_sb() drops the reference. * * But this ns can also have other tasks injected by setns()+fork(). @@ -268,7 +268,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns) */ for (;;) { set_current_state(TASK_INTERRUPTIBLE); - if (pid_ns->nr_hashed == init_pids) + if (pid_ns->pid_allocated == init_pids) break; schedule(); } -- cgit v1.3-8-gc7d7