summaryrefslogtreecommitdiffstats
path: root/lib/percpu_ida.c
diff options
context:
space:
mode:
authorKent Overstreet <kmo@daterainc.com>2013-08-16 22:04:37 +0000
committerNicholas Bellinger <nab@linux-iscsi.org>2013-09-09 14:29:15 -0700
commit798ab48eecdf659df9ae0064ca5c62626c651827 (patch)
tree1b78a050ec898f2647ad3c58c67a10462246740f /lib/percpu_ida.c
parent6faaa85f375543ea0d49a27e953ed18aec05ae56 (diff)
downloadlinux-798ab48eecdf659df9ae0064ca5c62626c651827.tar.gz
linux-798ab48eecdf659df9ae0064ca5c62626c651827.tar.xz
idr: Percpu ida
Percpu frontend for allocating ids. With percpu allocation (that works), it's impossible to guarantee it will always be possible to allocate all nr_tags - typically, some will be stuck on a remote percpu freelist where the current job can't get to them. We do guarantee that it will always be possible to allocate at least (nr_tags / 2) tags - this is done by keeping track of which and how many cpus have tags on their percpu freelists. On allocation failure if enough cpus have tags that there could potentially be (nr_tags / 2) tags stuck on remote percpu freelists, we then pick a remote cpu at random to steal from. Note that there's no cpu hotplug notifier - we don't care, because steal_tags() will eventually get the down cpu's tags. We _could_ satisfy more allocations if we had a notifier - but we'll still meet our guarantees and it's absolutely not a correctness issue, so I don't think it's worth the extra code. From akpm: "It looks OK to me (that's as close as I get to an ack :)) v6 changes: - Add #include <linux/cpumask.h> to include/linux/percpu_ida.h to make alpha/arc builds happy (Fengguang) - Move second (cpu >= nr_cpu_ids) check inside of first check scope in steal_tags() (akpm + nab) v5 changes: - Change percpu_ida->cpus_have_tags to cpumask_t (kmo + akpm) - Add comment for percpu_ida_cpu->lock + ->nr_free (kmo + akpm) - Convert steal_tags() to use cpumask_weight() + cpumask_next() + cpumask_first() + cpumask_clear_cpu() (kmo + akpm) - Add comment for alloc_global_tags() (kmo + akpm) - Convert percpu_ida_alloc() to use cpumask_set_cpu() (kmo + akpm) - Convert percpu_ida_free() to use cpumask_set_cpu() (kmo + akpm) - Drop percpu_ida->cpus_have_tags allocation in percpu_ida_init() (kmo + akpm) - Drop percpu_ida->cpus_have_tags kfree in percpu_ida_destroy() (kmo + akpm) - Add comment for percpu_ida_alloc @ gfp (kmo + akpm) - Move to percpu_ida.c + percpu_ida.h (kmo + akpm + nab) v4 changes: - Fix tags.c reference in percpu_ida_init (akpm) Signed-off-by: Kent Overstreet <kmo@daterainc.com> Cc: Tejun Heo <tj@kernel.org> Cc: Oleg Nesterov <oleg@redhat.com> Cc: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Jens Axboe <axboe@kernel.dk> Cc: "Nicholas A. Bellinger" <nab@linux-iscsi.org> Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Diffstat (limited to 'lib/percpu_ida.c')
-rw-r--r--lib/percpu_ida.c335
1 files changed, 335 insertions, 0 deletions
diff --git a/lib/percpu_ida.c b/lib/percpu_ida.c
new file mode 100644
index 000000000000..bab1ba2a4c71
--- /dev/null
+++ b/lib/percpu_ida.c
@@ -0,0 +1,335 @@
+/*
+ * Percpu IDA library
+ *
+ * Copyright (C) 2013 Datera, Inc. Kent Overstreet
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2, or (at
+ * your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ */
+
+#include <linux/bitmap.h>
+#include <linux/bitops.h>
+#include <linux/bug.h>
+#include <linux/err.h>
+#include <linux/export.h>
+#include <linux/hardirq.h>
+#include <linux/idr.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/spinlock.h>
+#include <linux/percpu_ida.h>
+
+/*
+ * Number of tags we move between the percpu freelist and the global freelist at
+ * a time
+ */
+#define IDA_PCPU_BATCH_MOVE 32U
+
+/* Max size of percpu freelist, */
+#define IDA_PCPU_SIZE ((IDA_PCPU_BATCH_MOVE * 3) / 2)
+
+struct percpu_ida_cpu {
+ /*
+ * Even though this is percpu, we need a lock for tag stealing by remote
+ * CPUs:
+ */
+ spinlock_t lock;
+
+ /* nr_free/freelist form a stack of free IDs */
+ unsigned nr_free;
+ unsigned freelist[];
+};
+
+static inline void move_tags(unsigned *dst, unsigned *dst_nr,
+ unsigned *src, unsigned *src_nr,
+ unsigned nr)
+{
+ *src_nr -= nr;
+ memcpy(dst + *dst_nr, src + *src_nr, sizeof(unsigned) * nr);
+ *dst_nr += nr;
+}
+
+/*
+ * Try to steal tags from a remote cpu's percpu freelist.
+ *
+ * We first check how many percpu freelists have tags - we don't steal tags
+ * unless enough percpu freelists have tags on them that it's possible more than
+ * half the total tags could be stuck on remote percpu freelists.
+ *
+ * Then we iterate through the cpus until we find some tags - we don't attempt
+ * to find the "best" cpu to steal from, to keep cacheline bouncing to a
+ * minimum.
+ */
+static inline void steal_tags(struct percpu_ida *pool,
+ struct percpu_ida_cpu *tags)
+{
+ unsigned cpus_have_tags, cpu = pool->cpu_last_stolen;
+ struct percpu_ida_cpu *remote;
+
+ for (cpus_have_tags = cpumask_weight(&pool->cpus_have_tags);
+ cpus_have_tags * IDA_PCPU_SIZE > pool->nr_tags / 2;
+ cpus_have_tags--) {
+ cpu = cpumask_next(cpu, &pool->cpus_have_tags);
+
+ if (cpu >= nr_cpu_ids) {
+ cpu = cpumask_first(&pool->cpus_have_tags);
+ if (cpu >= nr_cpu_ids)
+ BUG();
+ }
+
+ pool->cpu_last_stolen = cpu;
+ remote = per_cpu_ptr(pool->tag_cpu, cpu);
+
+ cpumask_clear_cpu(cpu, &pool->cpus_have_tags);
+
+ if (remote == tags)
+ continue;
+
+ spin_lock(&remote->lock);
+
+ if (remote->nr_free) {
+ memcpy(tags->freelist,
+ remote->freelist,
+ sizeof(unsigned) * remote->nr_free);
+
+ tags->nr_free = remote->nr_free;
+ remote->nr_free = 0;
+ }
+
+ spin_unlock(&remote->lock);
+
+ if (tags->nr_free)
+ break;
+ }
+}
+
+/*
+ * Pop up to IDA_PCPU_BATCH_MOVE IDs off the global freelist, and push them onto
+ * our percpu freelist:
+ */
+static inline void alloc_global_tags(struct percpu_ida *pool,
+ struct percpu_ida_cpu *tags)
+{
+ move_tags(tags->freelist, &tags->nr_free,
+ pool->freelist, &pool->nr_free,
+ min(pool->nr_free, IDA_PCPU_BATCH_MOVE));
+}
+
+static inline unsigned alloc_local_tag(struct percpu_ida *pool,
+ struct percpu_ida_cpu *tags)
+{
+ int tag = -ENOSPC;
+
+ spin_lock(&tags->lock);
+ if (tags->nr_free)
+ tag = tags->freelist[--tags->nr_free];
+ spin_unlock(&tags->lock);
+
+ return tag;
+}
+
+/**
+ * percpu_ida_alloc - allocate a tag
+ * @pool: pool to allocate from
+ * @gfp: gfp flags
+ *
+ * Returns a tag - an integer in the range [0..nr_tags) (passed to
+ * tag_pool_init()), or otherwise -ENOSPC on allocation failure.
+ *
+ * Safe to be called from interrupt context (assuming it isn't passed
+ * __GFP_WAIT, of course).
+ *
+ * @gfp indicates whether or not to wait until a free id is available (it's not
+ * used for internal memory allocations); thus if passed __GFP_WAIT we may sleep
+ * however long it takes until another thread frees an id (same semantics as a
+ * mempool).
+ *
+ * Will not fail if passed __GFP_WAIT.
+ */
+int percpu_ida_alloc(struct percpu_ida *pool, gfp_t gfp)
+{
+ DEFINE_WAIT(wait);
+ struct percpu_ida_cpu *tags;
+ unsigned long flags;
+ int tag;
+
+ local_irq_save(flags);
+ tags = this_cpu_ptr(pool->tag_cpu);
+
+ /* Fastpath */
+ tag = alloc_local_tag(pool, tags);
+ if (likely(tag >= 0)) {
+ local_irq_restore(flags);
+ return tag;
+ }
+
+ while (1) {
+ spin_lock(&pool->lock);
+
+ /*
+ * prepare_to_wait() must come before steal_tags(), in case
+ * percpu_ida_free() on another cpu flips a bit in
+ * cpus_have_tags
+ *
+ * global lock held and irqs disabled, don't need percpu lock
+ */
+ prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE);
+
+ if (!tags->nr_free)
+ alloc_global_tags(pool, tags);
+ if (!tags->nr_free)
+ steal_tags(pool, tags);
+
+ if (tags->nr_free) {
+ tag = tags->freelist[--tags->nr_free];
+ if (tags->nr_free)
+ cpumask_set_cpu(smp_processor_id(),
+ &pool->cpus_have_tags);
+ }
+
+ spin_unlock(&pool->lock);
+ local_irq_restore(flags);
+
+ if (tag >= 0 || !(gfp & __GFP_WAIT))
+ break;
+
+ schedule();
+
+ local_irq_save(flags);
+ tags = this_cpu_ptr(pool->tag_cpu);
+ }
+
+ finish_wait(&pool->wait, &wait);
+ return tag;
+}
+EXPORT_SYMBOL_GPL(percpu_ida_alloc);
+
+/**
+ * percpu_ida_free - free a tag
+ * @pool: pool @tag was allocated from
+ * @tag: a tag previously allocated with percpu_ida_alloc()
+ *
+ * Safe to be called from interrupt context.
+ */
+void percpu_ida_free(struct percpu_ida *pool, unsigned tag)
+{
+ struct percpu_ida_cpu *tags;
+ unsigned long flags;
+ unsigned nr_free;
+
+ BUG_ON(tag >= pool->nr_tags);
+
+ local_irq_save(flags);
+ tags = this_cpu_ptr(pool->tag_cpu);
+
+ spin_lock(&tags->lock);
+ tags->freelist[tags->nr_free++] = tag;
+
+ nr_free = tags->nr_free;
+ spin_unlock(&tags->lock);
+
+ if (nr_free == 1) {
+ cpumask_set_cpu(smp_processor_id(),
+ &pool->cpus_have_tags);
+ wake_up(&pool->wait);
+ }
+
+ if (nr_free == IDA_PCPU_SIZE) {
+ spin_lock(&pool->lock);
+
+ /*
+ * Global lock held and irqs disabled, don't need percpu
+ * lock
+ */
+ if (tags->nr_free == IDA_PCPU_SIZE) {
+ move_tags(pool->freelist, &pool->nr_free,
+ tags->freelist, &tags->nr_free,
+ IDA_PCPU_BATCH_MOVE);
+
+ wake_up(&pool->wait);
+ }
+ spin_unlock(&pool->lock);
+ }
+
+ local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(percpu_ida_free);
+
+/**
+ * percpu_ida_destroy - release a tag pool's resources
+ * @pool: pool to free
+ *
+ * Frees the resources allocated by percpu_ida_init().
+ */
+void percpu_ida_destroy(struct percpu_ida *pool)
+{
+ free_percpu(pool->tag_cpu);
+ free_pages((unsigned long) pool->freelist,
+ get_order(pool->nr_tags * sizeof(unsigned)));
+}
+EXPORT_SYMBOL_GPL(percpu_ida_destroy);
+
+/**
+ * percpu_ida_init - initialize a percpu tag pool
+ * @pool: pool to initialize
+ * @nr_tags: number of tags that will be available for allocation
+ *
+ * Initializes @pool so that it can be used to allocate tags - integers in the
+ * range [0, nr_tags). Typically, they'll be used by driver code to refer to a
+ * preallocated array of tag structures.
+ *
+ * Allocation is percpu, but sharding is limited by nr_tags - for best
+ * performance, the workload should not span more cpus than nr_tags / 128.
+ */
+int percpu_ida_init(struct percpu_ida *pool, unsigned long nr_tags)
+{
+ unsigned i, cpu, order;
+
+ memset(pool, 0, sizeof(*pool));
+
+ init_waitqueue_head(&pool->wait);
+ spin_lock_init(&pool->lock);
+ pool->nr_tags = nr_tags;
+
+ /* Guard against overflow */
+ if (nr_tags > (unsigned) INT_MAX + 1) {
+ pr_err("percpu_ida_init(): nr_tags too large\n");
+ return -EINVAL;
+ }
+
+ order = get_order(nr_tags * sizeof(unsigned));
+ pool->freelist = (void *) __get_free_pages(GFP_KERNEL, order);
+ if (!pool->freelist)
+ return -ENOMEM;
+
+ for (i = 0; i < nr_tags; i++)
+ pool->freelist[i] = i;
+
+ pool->nr_free = nr_tags;
+
+ pool->tag_cpu = __alloc_percpu(sizeof(struct percpu_ida_cpu) +
+ IDA_PCPU_SIZE * sizeof(unsigned),
+ sizeof(unsigned));
+ if (!pool->tag_cpu)
+ goto err;
+
+ for_each_possible_cpu(cpu)
+ spin_lock_init(&per_cpu_ptr(pool->tag_cpu, cpu)->lock);
+
+ return 0;
+err:
+ percpu_ida_destroy(pool);
+ return -ENOMEM;
+}
+EXPORT_SYMBOL_GPL(percpu_ida_init);