mirror of
https://github.com/edk2-porting/linux-next.git
synced 2024-12-27 14:43:58 +08:00
faf65dde84
The current implementation of the memcg accounting of the percpu memory is based on the idea of having two separate sets of chunks for accounted and non-accounted memory. This approach has an advantage of not wasting any extra memory for memcg data for non-accounted chunks, however it complicates the code and leads to a higher chunks number due to a lower chunk utilization. Instead of having two chunk types it's possible to declare all* chunks memcg-aware unless the kernel memory accounting is disabled globally by a boot option. The size of objcg_array is usually small in comparison to chunks themselves (it obviously depends on the number of CPUs), so even if some chunk will have no accounted allocations, the memory waste isn't significant and will likely be compensated by a higher chunk utilization. Also, with time more and more percpu allocations will likely become accounted. * The first chunk is initialized before the memory cgroup subsystem, so we don't know for sure whether we need to allocate obj_cgroups. Because it's small, let's make it free for use. Then we don't need to allocate obj_cgroups for it. Signed-off-by: Roman Gushchin <guro@fb.com> Signed-off-by: Dennis Zhou <dennis@kernel.org>
125 lines
3.0 KiB
C
125 lines
3.0 KiB
C
// SPDX-License-Identifier: GPL-2.0-only
|
|
/*
|
|
* mm/percpu-km.c - kernel memory based chunk allocation
|
|
*
|
|
* Copyright (C) 2010 SUSE Linux Products GmbH
|
|
* Copyright (C) 2010 Tejun Heo <tj@kernel.org>
|
|
*
|
|
* 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, gfp_t gfp)
|
|
{
|
|
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(gfp_t gfp)
|
|
{
|
|
const int nr_pages = pcpu_group_sizes[0] >> PAGE_SHIFT;
|
|
struct pcpu_chunk *chunk;
|
|
struct page *pages;
|
|
unsigned long flags;
|
|
int i;
|
|
|
|
chunk = pcpu_alloc_chunk(gfp);
|
|
if (!chunk)
|
|
return NULL;
|
|
|
|
pages = alloc_pages(gfp, 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);
|
|
|
|
spin_lock_irqsave(&pcpu_lock, flags);
|
|
pcpu_chunk_populated(chunk, 0, nr_pages);
|
|
spin_unlock_irqrestore(&pcpu_lock, flags);
|
|
|
|
pcpu_stats_chunk_alloc();
|
|
trace_percpu_create_chunk(chunk->base_addr);
|
|
|
|
return chunk;
|
|
}
|
|
|
|
static void pcpu_destroy_chunk(struct pcpu_chunk *chunk)
|
|
{
|
|
const int nr_pages = pcpu_group_sizes[0] >> PAGE_SHIFT;
|
|
|
|
if (!chunk)
|
|
return;
|
|
|
|
pcpu_stats_chunk_dealloc();
|
|
trace_percpu_destroy_chunk(chunk->base_addr);
|
|
|
|
if (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) {
|
|
pr_crit("can't handle more than one group\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)
|
|
pr_warn("wasting %zu pages per chunk\n",
|
|
alloc_pages - nr_pages);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static bool pcpu_should_reclaim_chunk(struct pcpu_chunk *chunk)
|
|
{
|
|
return false;
|
|
}
|