b38d08f318
At first, the percpu allocator required a sleepable context for both alloc and free paths and used pcpu_alloc_mutex to protect everything. Later, pcpu_lock was introduced to protect the index data structure so that the free path can be invoked from atomic contexts. The conversion only updated what's necessary and left most of the allocation path under pcpu_alloc_mutex. The percpu allocator is planned to add support for atomic allocation and this patch restructures locking so that the coverage of pcpu_alloc_mutex is further reduced. * pcpu_alloc() now grab pcpu_alloc_mutex only while creating a new chunk and populating the allocated area. Everything else is now protected soley by pcpu_lock. After this change, multiple instances of pcpu_extend_area_map() may race but the function already implements sufficient synchronization using pcpu_lock. This also allows multiple allocators to arrive at new chunk creation. To avoid creating multiple empty chunks back-to-back, a new chunk is created iff there is no other empty chunk after grabbing pcpu_alloc_mutex. * pcpu_lock is now held while modifying chunk->populated bitmap. After this, all data structures are protected by pcpu_lock. Signed-off-by: Tejun Heo <tj@kernel.org>
110 lines
2.8 KiB
C
110 lines
2.8 KiB
C
/*
|
|
* mm/percpu-km.c - kernel memory based chunk allocation
|
|
*
|
|
* Copyright (C) 2010 SUSE Linux Products GmbH
|
|
* Copyright (C) 2010 Tejun Heo <tj@kernel.org>
|
|
*
|
|
* This file is released under the GPLv2.
|
|
*
|
|
* Chunks are allocated as a contiguous kernel memory using gfp
|
|
* allocation. This is to be used on nommu architectures.
|
|
*
|
|
* To use percpu-km,
|
|
*
|
|
* - define CONFIG_NEED_PER_CPU_KM from the arch Kconfig.
|
|
*
|
|
* - CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK must not be defined. It's
|
|
* not compatible with PER_CPU_KM. EMBED_FIRST_CHUNK should work
|
|
* fine.
|
|
*
|
|
* - NUMA is not supported. When setting up the first chunk,
|
|
* @cpu_distance_fn should be NULL or report all CPUs to be nearer
|
|
* than or at LOCAL_DISTANCE.
|
|
*
|
|
* - It's best if the chunk size is power of two multiple of
|
|
* PAGE_SIZE. Because each chunk is allocated as a contiguous
|
|
* kernel memory block using alloc_pages(), memory will be wasted if
|
|
* chunk size is not aligned. percpu-km code will whine about it.
|
|
*/
|
|
|
|
#if defined(CONFIG_SMP) && defined(CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK)
|
|
#error "contiguous percpu allocation is incompatible with paged first chunk"
|
|
#endif
|
|
|
|
#include <linux/log2.h>
|
|
|
|
static int pcpu_populate_chunk(struct pcpu_chunk *chunk,
|
|
int page_start, int page_end)
|
|
{
|
|
return 0;
|
|
}
|
|
|
|
static void pcpu_depopulate_chunk(struct pcpu_chunk *chunk,
|
|
int page_start, int page_end)
|
|
{
|
|
/* nada */
|
|
}
|
|
|
|
static struct pcpu_chunk *pcpu_create_chunk(void)
|
|
{
|
|
const int nr_pages = pcpu_group_sizes[0] >> PAGE_SHIFT;
|
|
struct pcpu_chunk *chunk;
|
|
struct page *pages;
|
|
int i;
|
|
|
|
chunk = pcpu_alloc_chunk();
|
|
if (!chunk)
|
|
return NULL;
|
|
|
|
pages = alloc_pages(GFP_KERNEL, order_base_2(nr_pages));
|
|
if (!pages) {
|
|
pcpu_free_chunk(chunk);
|
|
return NULL;
|
|
}
|
|
|
|
for (i = 0; i < nr_pages; i++)
|
|
pcpu_set_page_chunk(nth_page(pages, i), chunk);
|
|
|
|
chunk->data = pages;
|
|
chunk->base_addr = page_address(pages) - pcpu_group_offsets[0];
|
|
|
|
spin_lock_irq(&pcpu_lock);
|
|
bitmap_fill(chunk->populated, nr_pages);
|
|
spin_unlock_irq(&pcpu_lock);
|
|
|
|
return chunk;
|
|
}
|
|
|
|
static void pcpu_destroy_chunk(struct pcpu_chunk *chunk)
|
|
{
|
|
const int nr_pages = pcpu_group_sizes[0] >> PAGE_SHIFT;
|
|
|
|
if (chunk && chunk->data)
|
|
__free_pages(chunk->data, order_base_2(nr_pages));
|
|
pcpu_free_chunk(chunk);
|
|
}
|
|
|
|
static struct page *pcpu_addr_to_page(void *addr)
|
|
{
|
|
return virt_to_page(addr);
|
|
}
|
|
|
|
static int __init pcpu_verify_alloc_info(const struct pcpu_alloc_info *ai)
|
|
{
|
|
size_t nr_pages, alloc_pages;
|
|
|
|
/* all units must be in a single group */
|
|
if (ai->nr_groups != 1) {
|
|
printk(KERN_CRIT "percpu: can't handle more than one groups\n");
|
|
return -EINVAL;
|
|
}
|
|
|
|
nr_pages = (ai->groups[0].nr_units * ai->unit_size) >> PAGE_SHIFT;
|
|
alloc_pages = roundup_pow_of_two(nr_pages);
|
|
|
|
if (alloc_pages > nr_pages)
|
|
printk(KERN_WARNING "percpu: wasting %zu pages per chunk\n",
|
|
alloc_pages - nr_pages);
|
|
|
|
return 0;
|
|
}
|