![Tejun Heo](/assets/img/avatar_default.png)
On extreme configuration (e.g. 32bit 32-way NUMA machine), lpage percpu first chunk allocator can consume too much of vmalloc space. Make it fall back to 4k allocator if the consumption goes over 20%. [ Impact: add sanity check for lpage percpu first chunk allocator ] Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Jan Beulich <JBeulich@novell.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Ingo Molnar <mingo@elte.hu>
547 lines
15 KiB
C
547 lines
15 KiB
C
#include <linux/kernel.h>
|
|
#include <linux/module.h>
|
|
#include <linux/init.h>
|
|
#include <linux/bootmem.h>
|
|
#include <linux/percpu.h>
|
|
#include <linux/kexec.h>
|
|
#include <linux/crash_dump.h>
|
|
#include <linux/smp.h>
|
|
#include <linux/topology.h>
|
|
#include <linux/pfn.h>
|
|
#include <asm/sections.h>
|
|
#include <asm/processor.h>
|
|
#include <asm/setup.h>
|
|
#include <asm/mpspec.h>
|
|
#include <asm/apicdef.h>
|
|
#include <asm/highmem.h>
|
|
#include <asm/proto.h>
|
|
#include <asm/cpumask.h>
|
|
#include <asm/cpu.h>
|
|
#include <asm/stackprotector.h>
|
|
|
|
#ifdef CONFIG_DEBUG_PER_CPU_MAPS
|
|
# define DBG(x...) printk(KERN_DEBUG x)
|
|
#else
|
|
# define DBG(x...)
|
|
#endif
|
|
|
|
DEFINE_PER_CPU(int, cpu_number);
|
|
EXPORT_PER_CPU_SYMBOL(cpu_number);
|
|
|
|
#ifdef CONFIG_X86_64
|
|
#define BOOT_PERCPU_OFFSET ((unsigned long)__per_cpu_load)
|
|
#else
|
|
#define BOOT_PERCPU_OFFSET 0
|
|
#endif
|
|
|
|
DEFINE_PER_CPU(unsigned long, this_cpu_off) = BOOT_PERCPU_OFFSET;
|
|
EXPORT_PER_CPU_SYMBOL(this_cpu_off);
|
|
|
|
unsigned long __per_cpu_offset[NR_CPUS] __read_mostly = {
|
|
[0 ... NR_CPUS-1] = BOOT_PERCPU_OFFSET,
|
|
};
|
|
EXPORT_SYMBOL(__per_cpu_offset);
|
|
|
|
/*
|
|
* On x86_64 symbols referenced from code should be reachable using
|
|
* 32bit relocations. Reserve space for static percpu variables in
|
|
* modules so that they are always served from the first chunk which
|
|
* is located at the percpu segment base. On x86_32, anything can
|
|
* address anywhere. No need to reserve space in the first chunk.
|
|
*/
|
|
#ifdef CONFIG_X86_64
|
|
#define PERCPU_FIRST_CHUNK_RESERVE PERCPU_MODULE_RESERVE
|
|
#else
|
|
#define PERCPU_FIRST_CHUNK_RESERVE 0
|
|
#endif
|
|
|
|
/**
|
|
* pcpu_need_numa - determine percpu allocation needs to consider NUMA
|
|
*
|
|
* If NUMA is not configured or there is only one NUMA node available,
|
|
* there is no reason to consider NUMA. This function determines
|
|
* whether percpu allocation should consider NUMA or not.
|
|
*
|
|
* RETURNS:
|
|
* true if NUMA should be considered; otherwise, false.
|
|
*/
|
|
static bool __init pcpu_need_numa(void)
|
|
{
|
|
#ifdef CONFIG_NEED_MULTIPLE_NODES
|
|
pg_data_t *last = NULL;
|
|
unsigned int cpu;
|
|
|
|
for_each_possible_cpu(cpu) {
|
|
int node = early_cpu_to_node(cpu);
|
|
|
|
if (node_online(node) && NODE_DATA(node) &&
|
|
last && last != NODE_DATA(node))
|
|
return true;
|
|
|
|
last = NODE_DATA(node);
|
|
}
|
|
#endif
|
|
return false;
|
|
}
|
|
|
|
/**
|
|
* pcpu_alloc_bootmem - NUMA friendly alloc_bootmem wrapper for percpu
|
|
* @cpu: cpu to allocate for
|
|
* @size: size allocation in bytes
|
|
* @align: alignment
|
|
*
|
|
* Allocate @size bytes aligned at @align for cpu @cpu. This wrapper
|
|
* does the right thing for NUMA regardless of the current
|
|
* configuration.
|
|
*
|
|
* RETURNS:
|
|
* Pointer to the allocated area on success, NULL on failure.
|
|
*/
|
|
static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size,
|
|
unsigned long align)
|
|
{
|
|
const unsigned long goal = __pa(MAX_DMA_ADDRESS);
|
|
#ifdef CONFIG_NEED_MULTIPLE_NODES
|
|
int node = early_cpu_to_node(cpu);
|
|
void *ptr;
|
|
|
|
if (!node_online(node) || !NODE_DATA(node)) {
|
|
ptr = __alloc_bootmem_nopanic(size, align, goal);
|
|
pr_info("cpu %d has no node %d or node-local memory\n",
|
|
cpu, node);
|
|
pr_debug("per cpu data for cpu%d %lu bytes at %016lx\n",
|
|
cpu, size, __pa(ptr));
|
|
} else {
|
|
ptr = __alloc_bootmem_node_nopanic(NODE_DATA(node),
|
|
size, align, goal);
|
|
pr_debug("per cpu data for cpu%d %lu bytes on node%d at "
|
|
"%016lx\n", cpu, size, node, __pa(ptr));
|
|
}
|
|
return ptr;
|
|
#else
|
|
return __alloc_bootmem_nopanic(size, align, goal);
|
|
#endif
|
|
}
|
|
|
|
/*
|
|
* Large page remap allocator
|
|
*
|
|
* This allocator uses PMD page as unit. A PMD page is allocated for
|
|
* each cpu and each is remapped into vmalloc area using PMD mapping.
|
|
* As PMD page is quite large, only part of it is used for the first
|
|
* chunk. Unused part is returned to the bootmem allocator.
|
|
*
|
|
* So, the PMD pages are mapped twice - once to the physical mapping
|
|
* and to the vmalloc area for the first percpu chunk. The double
|
|
* mapping does add one more PMD TLB entry pressure but still is much
|
|
* better than only using 4k mappings while still being NUMA friendly.
|
|
*/
|
|
#ifdef CONFIG_NEED_MULTIPLE_NODES
|
|
struct pcpul_ent {
|
|
unsigned int cpu;
|
|
void *ptr;
|
|
};
|
|
|
|
static size_t pcpul_size;
|
|
static struct pcpul_ent *pcpul_map;
|
|
static struct vm_struct pcpul_vm;
|
|
|
|
static struct page * __init pcpul_get_page(unsigned int cpu, int pageno)
|
|
{
|
|
size_t off = (size_t)pageno << PAGE_SHIFT;
|
|
|
|
if (off >= pcpul_size)
|
|
return NULL;
|
|
|
|
return virt_to_page(pcpul_map[cpu].ptr + off);
|
|
}
|
|
|
|
static ssize_t __init setup_pcpu_lpage(size_t static_size, bool chosen)
|
|
{
|
|
size_t map_size, dyn_size;
|
|
unsigned int cpu;
|
|
int i, j;
|
|
ssize_t ret;
|
|
|
|
if (!chosen) {
|
|
size_t vm_size = VMALLOC_END - VMALLOC_START;
|
|
size_t tot_size = num_possible_cpus() * PMD_SIZE;
|
|
|
|
/* on non-NUMA, embedding is better */
|
|
if (!pcpu_need_numa())
|
|
return -EINVAL;
|
|
|
|
/* don't consume more than 20% of vmalloc area */
|
|
if (tot_size > vm_size / 5) {
|
|
pr_info("PERCPU: too large chunk size %zuMB for "
|
|
"large page remap\n", tot_size >> 20);
|
|
return -EINVAL;
|
|
}
|
|
}
|
|
|
|
/* need PSE */
|
|
if (!cpu_has_pse) {
|
|
pr_warning("PERCPU: lpage allocator requires PSE\n");
|
|
return -EINVAL;
|
|
}
|
|
|
|
/*
|
|
* Currently supports only single page. Supporting multiple
|
|
* pages won't be too difficult if it ever becomes necessary.
|
|
*/
|
|
pcpul_size = PFN_ALIGN(static_size + PERCPU_MODULE_RESERVE +
|
|
PERCPU_DYNAMIC_RESERVE);
|
|
if (pcpul_size > PMD_SIZE) {
|
|
pr_warning("PERCPU: static data is larger than large page, "
|
|
"can't use large page\n");
|
|
return -EINVAL;
|
|
}
|
|
dyn_size = pcpul_size - static_size - PERCPU_FIRST_CHUNK_RESERVE;
|
|
|
|
/* allocate pointer array and alloc large pages */
|
|
map_size = PFN_ALIGN(num_possible_cpus() * sizeof(pcpul_map[0]));
|
|
pcpul_map = alloc_bootmem(map_size);
|
|
|
|
for_each_possible_cpu(cpu) {
|
|
pcpul_map[cpu].cpu = cpu;
|
|
pcpul_map[cpu].ptr = pcpu_alloc_bootmem(cpu, PMD_SIZE,
|
|
PMD_SIZE);
|
|
if (!pcpul_map[cpu].ptr) {
|
|
pr_warning("PERCPU: failed to allocate large page "
|
|
"for cpu%u\n", cpu);
|
|
goto enomem;
|
|
}
|
|
|
|
/*
|
|
* Only use pcpul_size bytes and give back the rest.
|
|
*
|
|
* Ingo: The 2MB up-rounding bootmem is needed to make
|
|
* sure the partial 2MB page is still fully RAM - it's
|
|
* not well-specified to have a PAT-incompatible area
|
|
* (unmapped RAM, device memory, etc.) in that hole.
|
|
*/
|
|
free_bootmem(__pa(pcpul_map[cpu].ptr + pcpul_size),
|
|
PMD_SIZE - pcpul_size);
|
|
|
|
memcpy(pcpul_map[cpu].ptr, __per_cpu_load, static_size);
|
|
}
|
|
|
|
/* allocate address and map */
|
|
pcpul_vm.flags = VM_ALLOC;
|
|
pcpul_vm.size = num_possible_cpus() * PMD_SIZE;
|
|
vm_area_register_early(&pcpul_vm, PMD_SIZE);
|
|
|
|
for_each_possible_cpu(cpu) {
|
|
pmd_t *pmd, pmd_v;
|
|
|
|
pmd = populate_extra_pmd((unsigned long)pcpul_vm.addr +
|
|
cpu * PMD_SIZE);
|
|
pmd_v = pfn_pmd(page_to_pfn(virt_to_page(pcpul_map[cpu].ptr)),
|
|
PAGE_KERNEL_LARGE);
|
|
set_pmd(pmd, pmd_v);
|
|
}
|
|
|
|
/* we're ready, commit */
|
|
pr_info("PERCPU: Remapped at %p with large pages, static data "
|
|
"%zu bytes\n", pcpul_vm.addr, static_size);
|
|
|
|
ret = pcpu_setup_first_chunk(pcpul_get_page, static_size,
|
|
PERCPU_FIRST_CHUNK_RESERVE, dyn_size,
|
|
PMD_SIZE, pcpul_vm.addr, NULL);
|
|
|
|
/* sort pcpul_map array for pcpu_lpage_remapped() */
|
|
for (i = 0; i < num_possible_cpus() - 1; i++)
|
|
for (j = i + 1; j < num_possible_cpus(); j++)
|
|
if (pcpul_map[i].ptr > pcpul_map[j].ptr) {
|
|
struct pcpul_ent tmp = pcpul_map[i];
|
|
pcpul_map[i] = pcpul_map[j];
|
|
pcpul_map[j] = tmp;
|
|
}
|
|
|
|
return ret;
|
|
|
|
enomem:
|
|
for_each_possible_cpu(cpu)
|
|
if (pcpul_map[cpu].ptr)
|
|
free_bootmem(__pa(pcpul_map[cpu].ptr), pcpul_size);
|
|
free_bootmem(__pa(pcpul_map), map_size);
|
|
return -ENOMEM;
|
|
}
|
|
|
|
/**
|
|
* pcpu_lpage_remapped - determine whether a kaddr is in pcpul recycled area
|
|
* @kaddr: the kernel address in question
|
|
*
|
|
* Determine whether @kaddr falls in the pcpul recycled area. This is
|
|
* used by pageattr to detect VM aliases and break up the pcpu PMD
|
|
* mapping such that the same physical page is not mapped under
|
|
* different attributes.
|
|
*
|
|
* The recycled area is always at the tail of a partially used PMD
|
|
* page.
|
|
*
|
|
* RETURNS:
|
|
* Address of corresponding remapped pcpu address if match is found;
|
|
* otherwise, NULL.
|
|
*/
|
|
void *pcpu_lpage_remapped(void *kaddr)
|
|
{
|
|
void *pmd_addr = (void *)((unsigned long)kaddr & PMD_MASK);
|
|
unsigned long offset = (unsigned long)kaddr & ~PMD_MASK;
|
|
int left = 0, right = num_possible_cpus() - 1;
|
|
int pos;
|
|
|
|
/* pcpul in use at all? */
|
|
if (!pcpul_map)
|
|
return NULL;
|
|
|
|
/* okay, perform binary search */
|
|
while (left <= right) {
|
|
pos = (left + right) / 2;
|
|
|
|
if (pcpul_map[pos].ptr < pmd_addr)
|
|
left = pos + 1;
|
|
else if (pcpul_map[pos].ptr > pmd_addr)
|
|
right = pos - 1;
|
|
else {
|
|
/* it shouldn't be in the area for the first chunk */
|
|
WARN_ON(offset < pcpul_size);
|
|
|
|
return pcpul_vm.addr +
|
|
pcpul_map[pos].cpu * PMD_SIZE + offset;
|
|
}
|
|
}
|
|
|
|
return NULL;
|
|
}
|
|
#else
|
|
static ssize_t __init setup_pcpu_lpage(size_t static_size, bool chosen)
|
|
{
|
|
return -EINVAL;
|
|
}
|
|
#endif
|
|
|
|
/*
|
|
* Embedding allocator
|
|
*
|
|
* The first chunk is sized to just contain the static area plus
|
|
* module and dynamic reserves and embedded into linear physical
|
|
* mapping so that it can use PMD mapping without additional TLB
|
|
* pressure.
|
|
*/
|
|
static ssize_t __init setup_pcpu_embed(size_t static_size, bool chosen)
|
|
{
|
|
size_t reserve = PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE;
|
|
|
|
/*
|
|
* If large page isn't supported, there's no benefit in doing
|
|
* this. Also, embedding allocation doesn't play well with
|
|
* NUMA.
|
|
*/
|
|
if (!chosen && (!cpu_has_pse || pcpu_need_numa()))
|
|
return -EINVAL;
|
|
|
|
return pcpu_embed_first_chunk(static_size, PERCPU_FIRST_CHUNK_RESERVE,
|
|
reserve - PERCPU_FIRST_CHUNK_RESERVE, -1);
|
|
}
|
|
|
|
/*
|
|
* 4k page allocator
|
|
*
|
|
* This is the basic allocator. Static percpu area is allocated
|
|
* page-by-page and most of initialization is done by the generic
|
|
* setup function.
|
|
*/
|
|
static struct page **pcpu4k_pages __initdata;
|
|
static int pcpu4k_nr_static_pages __initdata;
|
|
|
|
static struct page * __init pcpu4k_get_page(unsigned int cpu, int pageno)
|
|
{
|
|
if (pageno < pcpu4k_nr_static_pages)
|
|
return pcpu4k_pages[cpu * pcpu4k_nr_static_pages + pageno];
|
|
return NULL;
|
|
}
|
|
|
|
static void __init pcpu4k_populate_pte(unsigned long addr)
|
|
{
|
|
populate_extra_pte(addr);
|
|
}
|
|
|
|
static ssize_t __init setup_pcpu_4k(size_t static_size)
|
|
{
|
|
size_t pages_size;
|
|
unsigned int cpu;
|
|
int i, j;
|
|
ssize_t ret;
|
|
|
|
pcpu4k_nr_static_pages = PFN_UP(static_size);
|
|
|
|
/* unaligned allocations can't be freed, round up to page size */
|
|
pages_size = PFN_ALIGN(pcpu4k_nr_static_pages * num_possible_cpus()
|
|
* sizeof(pcpu4k_pages[0]));
|
|
pcpu4k_pages = alloc_bootmem(pages_size);
|
|
|
|
/* allocate and copy */
|
|
j = 0;
|
|
for_each_possible_cpu(cpu)
|
|
for (i = 0; i < pcpu4k_nr_static_pages; i++) {
|
|
void *ptr;
|
|
|
|
ptr = pcpu_alloc_bootmem(cpu, PAGE_SIZE, PAGE_SIZE);
|
|
if (!ptr) {
|
|
pr_warning("PERCPU: failed to allocate "
|
|
"4k page for cpu%u\n", cpu);
|
|
goto enomem;
|
|
}
|
|
|
|
memcpy(ptr, __per_cpu_load + i * PAGE_SIZE, PAGE_SIZE);
|
|
pcpu4k_pages[j++] = virt_to_page(ptr);
|
|
}
|
|
|
|
/* we're ready, commit */
|
|
pr_info("PERCPU: Allocated %d 4k pages, static data %zu bytes\n",
|
|
pcpu4k_nr_static_pages, static_size);
|
|
|
|
ret = pcpu_setup_first_chunk(pcpu4k_get_page, static_size,
|
|
PERCPU_FIRST_CHUNK_RESERVE, -1,
|
|
-1, NULL, pcpu4k_populate_pte);
|
|
goto out_free_ar;
|
|
|
|
enomem:
|
|
while (--j >= 0)
|
|
free_bootmem(__pa(page_address(pcpu4k_pages[j])), PAGE_SIZE);
|
|
ret = -ENOMEM;
|
|
out_free_ar:
|
|
free_bootmem(__pa(pcpu4k_pages), pages_size);
|
|
return ret;
|
|
}
|
|
|
|
/* for explicit first chunk allocator selection */
|
|
static char pcpu_chosen_alloc[16] __initdata;
|
|
|
|
static int __init percpu_alloc_setup(char *str)
|
|
{
|
|
strncpy(pcpu_chosen_alloc, str, sizeof(pcpu_chosen_alloc) - 1);
|
|
return 0;
|
|
}
|
|
early_param("percpu_alloc", percpu_alloc_setup);
|
|
|
|
static inline void setup_percpu_segment(int cpu)
|
|
{
|
|
#ifdef CONFIG_X86_32
|
|
struct desc_struct gdt;
|
|
|
|
pack_descriptor(&gdt, per_cpu_offset(cpu), 0xFFFFF,
|
|
0x2 | DESCTYPE_S, 0x8);
|
|
gdt.s = 1;
|
|
write_gdt_entry(get_cpu_gdt_table(cpu),
|
|
GDT_ENTRY_PERCPU, &gdt, DESCTYPE_S);
|
|
#endif
|
|
}
|
|
|
|
void __init setup_per_cpu_areas(void)
|
|
{
|
|
size_t static_size = __per_cpu_end - __per_cpu_start;
|
|
unsigned int cpu;
|
|
unsigned long delta;
|
|
size_t pcpu_unit_size;
|
|
ssize_t ret;
|
|
|
|
pr_info("NR_CPUS:%d nr_cpumask_bits:%d nr_cpu_ids:%d nr_node_ids:%d\n",
|
|
NR_CPUS, nr_cpumask_bits, nr_cpu_ids, nr_node_ids);
|
|
|
|
/*
|
|
* Allocate percpu area. If PSE is supported, try to make use
|
|
* of large page mappings. Please read comments on top of
|
|
* each allocator for details.
|
|
*/
|
|
ret = -EINVAL;
|
|
if (strlen(pcpu_chosen_alloc)) {
|
|
if (strcmp(pcpu_chosen_alloc, "4k")) {
|
|
if (!strcmp(pcpu_chosen_alloc, "lpage"))
|
|
ret = setup_pcpu_lpage(static_size, true);
|
|
else if (!strcmp(pcpu_chosen_alloc, "embed"))
|
|
ret = setup_pcpu_embed(static_size, true);
|
|
else
|
|
pr_warning("PERCPU: unknown allocator %s "
|
|
"specified\n", pcpu_chosen_alloc);
|
|
if (ret < 0)
|
|
pr_warning("PERCPU: %s allocator failed (%zd), "
|
|
"falling back to 4k\n",
|
|
pcpu_chosen_alloc, ret);
|
|
}
|
|
} else {
|
|
ret = setup_pcpu_lpage(static_size, false);
|
|
if (ret < 0)
|
|
ret = setup_pcpu_embed(static_size, false);
|
|
}
|
|
if (ret < 0)
|
|
ret = setup_pcpu_4k(static_size);
|
|
if (ret < 0)
|
|
panic("cannot allocate static percpu area (%zu bytes, err=%zd)",
|
|
static_size, ret);
|
|
|
|
pcpu_unit_size = ret;
|
|
|
|
/* alrighty, percpu areas up and running */
|
|
delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start;
|
|
for_each_possible_cpu(cpu) {
|
|
per_cpu_offset(cpu) = delta + cpu * pcpu_unit_size;
|
|
per_cpu(this_cpu_off, cpu) = per_cpu_offset(cpu);
|
|
per_cpu(cpu_number, cpu) = cpu;
|
|
setup_percpu_segment(cpu);
|
|
setup_stack_canary_segment(cpu);
|
|
/*
|
|
* Copy data used in early init routines from the
|
|
* initial arrays to the per cpu data areas. These
|
|
* arrays then become expendable and the *_early_ptr's
|
|
* are zeroed indicating that the static arrays are
|
|
* gone.
|
|
*/
|
|
#ifdef CONFIG_X86_LOCAL_APIC
|
|
per_cpu(x86_cpu_to_apicid, cpu) =
|
|
early_per_cpu_map(x86_cpu_to_apicid, cpu);
|
|
per_cpu(x86_bios_cpu_apicid, cpu) =
|
|
early_per_cpu_map(x86_bios_cpu_apicid, cpu);
|
|
#endif
|
|
#ifdef CONFIG_X86_64
|
|
per_cpu(irq_stack_ptr, cpu) =
|
|
per_cpu(irq_stack_union.irq_stack, cpu) +
|
|
IRQ_STACK_SIZE - 64;
|
|
#ifdef CONFIG_NUMA
|
|
per_cpu(x86_cpu_to_node_map, cpu) =
|
|
early_per_cpu_map(x86_cpu_to_node_map, cpu);
|
|
#endif
|
|
#endif
|
|
/*
|
|
* Up to this point, the boot CPU has been using .data.init
|
|
* area. Reload any changed state for the boot CPU.
|
|
*/
|
|
if (cpu == boot_cpu_id)
|
|
switch_to_new_gdt(cpu);
|
|
}
|
|
|
|
/* indicate the early static arrays will soon be gone */
|
|
#ifdef CONFIG_X86_LOCAL_APIC
|
|
early_per_cpu_ptr(x86_cpu_to_apicid) = NULL;
|
|
early_per_cpu_ptr(x86_bios_cpu_apicid) = NULL;
|
|
#endif
|
|
#if defined(CONFIG_X86_64) && defined(CONFIG_NUMA)
|
|
early_per_cpu_ptr(x86_cpu_to_node_map) = NULL;
|
|
#endif
|
|
|
|
#if defined(CONFIG_X86_64) && defined(CONFIG_NUMA)
|
|
/*
|
|
* make sure boot cpu node_number is right, when boot cpu is on the
|
|
* node that doesn't have mem installed
|
|
*/
|
|
per_cpu(node_number, boot_cpu_id) = cpu_to_node(boot_cpu_id);
|
|
#endif
|
|
|
|
/* Setup node to cpumask map */
|
|
setup_node_to_cpumask_map();
|
|
|
|
/* Setup cpu initialized, callin, callout masks */
|
|
setup_cpu_local_masks();
|
|
}
|