linux_dsm_epyc7002/arch/mips/mm/init.c
David Daney 43064c0c8e MIPS: Handle initmem in systems with kernel not in add_memory_region() mem
This patch addresses a couple of related problems:

1) The kernel may reside in physical memory outside of the ranges set
   by plat_mem_setup().  If this is the case, init mem cannot be
   reused as it resides outside of the range of pages that the kernel
   memory allocators control.

2) initrd images might be loaded in physical memory outside of the
   ranges set by plat_mem_setup().  The memory likewise cannot be
   reused.  The patch doesn't handle this specific case, but the
   infrastructure is useful for future patches that do.

The crux of the problem is that there are memory regions that need be
memory_present(), but that cannot be free_bootmem() at the time of
arch_mem_init().  We create a new type of memory (BOOT_MEM_INIT_RAM)
for use with add_memory_region().  Then arch_mem_init() adds the init
mem with this type if the init mem is not already covered by existing
ranges.

When memory is being freed into the bootmem allocator, we skip the
BOOT_MEM_INIT_RAM ranges so they are not clobbered, but we do signal
them as memory_present().  This way when they are later freed, the
necessary memory manager structures have initialized and the Sparse
allocater is prevented from crashing.

The Octeon specific code that handled this case is removed, because
the new general purpose code handles the case.

Signed-off-by: David Daney <ddaney@caviumnetworks.com>
To: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/1988/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
2011-12-07 22:03:45 +00:00

488 lines
12 KiB
C

/*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* Copyright (C) 1994 - 2000 Ralf Baechle
* Copyright (C) 1999, 2000 Silicon Graphics, Inc.
* Kevin D. Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
* Copyright (C) 2000 MIPS Technologies, Inc. All rights reserved.
*/
#include <linux/bug.h>
#include <linux/init.h>
#include <linux/module.h>
#include <linux/signal.h>
#include <linux/sched.h>
#include <linux/smp.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/types.h>
#include <linux/pagemap.h>
#include <linux/ptrace.h>
#include <linux/mman.h>
#include <linux/mm.h>
#include <linux/bootmem.h>
#include <linux/highmem.h>
#include <linux/swap.h>
#include <linux/proc_fs.h>
#include <linux/pfn.h>
#include <linux/hardirq.h>
#include <linux/gfp.h>
#include <asm/asm-offsets.h>
#include <asm/bootinfo.h>
#include <asm/cachectl.h>
#include <asm/cpu.h>
#include <asm/dma.h>
#include <asm/kmap_types.h>
#include <asm/mmu_context.h>
#include <asm/sections.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/tlb.h>
#include <asm/fixmap.h>
/* Atomicity and interruptability */
#ifdef CONFIG_MIPS_MT_SMTC
#include <asm/mipsmtregs.h>
#define ENTER_CRITICAL(flags) \
{ \
unsigned int mvpflags; \
local_irq_save(flags);\
mvpflags = dvpe()
#define EXIT_CRITICAL(flags) \
evpe(mvpflags); \
local_irq_restore(flags); \
}
#else
#define ENTER_CRITICAL(flags) local_irq_save(flags)
#define EXIT_CRITICAL(flags) local_irq_restore(flags)
#endif /* CONFIG_MIPS_MT_SMTC */
/*
* We have up to 8 empty zeroed pages so we can map one of the right colour
* when needed. This is necessary only on R4000 / R4400 SC and MC versions
* where we have to avoid VCED / VECI exceptions for good performance at
* any price. Since page is never written to after the initialization we
* don't have to care about aliases on other CPUs.
*/
unsigned long empty_zero_page, zero_page_mask;
EXPORT_SYMBOL_GPL(empty_zero_page);
/*
* Not static inline because used by IP27 special magic initialization code
*/
unsigned long setup_zero_pages(void)
{
unsigned int order;
unsigned long size;
struct page *page;
if (cpu_has_vce)
order = 3;
else
order = 0;
empty_zero_page = __get_free_pages(GFP_KERNEL | __GFP_ZERO, order);
if (!empty_zero_page)
panic("Oh boy, that early out of memory?");
page = virt_to_page((void *)empty_zero_page);
split_page(page, order);
while (page < virt_to_page((void *)(empty_zero_page + (PAGE_SIZE << order)))) {
SetPageReserved(page);
page++;
}
size = PAGE_SIZE << order;
zero_page_mask = (size - 1) & PAGE_MASK;
return 1UL << order;
}
#ifdef CONFIG_MIPS_MT_SMTC
static pte_t *kmap_coherent_pte;
static void __init kmap_coherent_init(void)
{
unsigned long vaddr;
/* cache the first coherent kmap pte */
vaddr = __fix_to_virt(FIX_CMAP_BEGIN);
kmap_coherent_pte = kmap_get_fixmap_pte(vaddr);
}
#else
static inline void kmap_coherent_init(void) {}
#endif
void *kmap_coherent(struct page *page, unsigned long addr)
{
enum fixed_addresses idx;
unsigned long vaddr, flags, entrylo;
unsigned long old_ctx;
pte_t pte;
int tlbidx;
BUG_ON(Page_dcache_dirty(page));
inc_preempt_count();
idx = (addr >> PAGE_SHIFT) & (FIX_N_COLOURS - 1);
#ifdef CONFIG_MIPS_MT_SMTC
idx += FIX_N_COLOURS * smp_processor_id() +
(in_interrupt() ? (FIX_N_COLOURS * NR_CPUS) : 0);
#else
idx += in_interrupt() ? FIX_N_COLOURS : 0;
#endif
vaddr = __fix_to_virt(FIX_CMAP_END - idx);
pte = mk_pte(page, PAGE_KERNEL);
#if defined(CONFIG_64BIT_PHYS_ADDR) && defined(CONFIG_CPU_MIPS32)
entrylo = pte.pte_high;
#else
entrylo = pte_to_entrylo(pte_val(pte));
#endif
ENTER_CRITICAL(flags);
old_ctx = read_c0_entryhi();
write_c0_entryhi(vaddr & (PAGE_MASK << 1));
write_c0_entrylo0(entrylo);
write_c0_entrylo1(entrylo);
#ifdef CONFIG_MIPS_MT_SMTC
set_pte(kmap_coherent_pte - (FIX_CMAP_END - idx), pte);
/* preload TLB instead of local_flush_tlb_one() */
mtc0_tlbw_hazard();
tlb_probe();
tlb_probe_hazard();
tlbidx = read_c0_index();
mtc0_tlbw_hazard();
if (tlbidx < 0)
tlb_write_random();
else
tlb_write_indexed();
#else
tlbidx = read_c0_wired();
write_c0_wired(tlbidx + 1);
write_c0_index(tlbidx);
mtc0_tlbw_hazard();
tlb_write_indexed();
#endif
tlbw_use_hazard();
write_c0_entryhi(old_ctx);
EXIT_CRITICAL(flags);
return (void*) vaddr;
}
#define UNIQUE_ENTRYHI(idx) (CKSEG0 + ((idx) << (PAGE_SHIFT + 1)))
void kunmap_coherent(void)
{
#ifndef CONFIG_MIPS_MT_SMTC
unsigned int wired;
unsigned long flags, old_ctx;
ENTER_CRITICAL(flags);
old_ctx = read_c0_entryhi();
wired = read_c0_wired() - 1;
write_c0_wired(wired);
write_c0_index(wired);
write_c0_entryhi(UNIQUE_ENTRYHI(wired));
write_c0_entrylo0(0);
write_c0_entrylo1(0);
mtc0_tlbw_hazard();
tlb_write_indexed();
tlbw_use_hazard();
write_c0_entryhi(old_ctx);
EXIT_CRITICAL(flags);
#endif
dec_preempt_count();
preempt_check_resched();
}
void copy_user_highpage(struct page *to, struct page *from,
unsigned long vaddr, struct vm_area_struct *vma)
{
void *vfrom, *vto;
vto = kmap_atomic(to, KM_USER1);
if (cpu_has_dc_aliases &&
page_mapped(from) && !Page_dcache_dirty(from)) {
vfrom = kmap_coherent(from, vaddr);
copy_page(vto, vfrom);
kunmap_coherent();
} else {
vfrom = kmap_atomic(from, KM_USER0);
copy_page(vto, vfrom);
kunmap_atomic(vfrom, KM_USER0);
}
if ((!cpu_has_ic_fills_f_dc) ||
pages_do_alias((unsigned long)vto, vaddr & PAGE_MASK))
flush_data_cache_page((unsigned long)vto);
kunmap_atomic(vto, KM_USER1);
/* Make sure this page is cleared on other CPU's too before using it */
smp_wmb();
}
void copy_to_user_page(struct vm_area_struct *vma,
struct page *page, unsigned long vaddr, void *dst, const void *src,
unsigned long len)
{
if (cpu_has_dc_aliases &&
page_mapped(page) && !Page_dcache_dirty(page)) {
void *vto = kmap_coherent(page, vaddr) + (vaddr & ~PAGE_MASK);
memcpy(vto, src, len);
kunmap_coherent();
} else {
memcpy(dst, src, len);
if (cpu_has_dc_aliases)
SetPageDcacheDirty(page);
}
if ((vma->vm_flags & VM_EXEC) && !cpu_has_ic_fills_f_dc)
flush_cache_page(vma, vaddr, page_to_pfn(page));
}
void copy_from_user_page(struct vm_area_struct *vma,
struct page *page, unsigned long vaddr, void *dst, const void *src,
unsigned long len)
{
if (cpu_has_dc_aliases &&
page_mapped(page) && !Page_dcache_dirty(page)) {
void *vfrom = kmap_coherent(page, vaddr) + (vaddr & ~PAGE_MASK);
memcpy(dst, vfrom, len);
kunmap_coherent();
} else {
memcpy(dst, src, len);
if (cpu_has_dc_aliases)
SetPageDcacheDirty(page);
}
}
void __init fixrange_init(unsigned long start, unsigned long end,
pgd_t *pgd_base)
{
#if defined(CONFIG_HIGHMEM) || defined(CONFIG_MIPS_MT_SMTC)
pgd_t *pgd;
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
int i, j, k;
unsigned long vaddr;
vaddr = start;
i = __pgd_offset(vaddr);
j = __pud_offset(vaddr);
k = __pmd_offset(vaddr);
pgd = pgd_base + i;
for ( ; (i < PTRS_PER_PGD) && (vaddr < end); pgd++, i++) {
pud = (pud_t *)pgd;
for ( ; (j < PTRS_PER_PUD) && (vaddr < end); pud++, j++) {
pmd = (pmd_t *)pud;
for (; (k < PTRS_PER_PMD) && (vaddr < end); pmd++, k++) {
if (pmd_none(*pmd)) {
pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
set_pmd(pmd, __pmd((unsigned long)pte));
BUG_ON(pte != pte_offset_kernel(pmd, 0));
}
vaddr += PMD_SIZE;
}
k = 0;
}
j = 0;
}
#endif
}
#ifndef CONFIG_NEED_MULTIPLE_NODES
int page_is_ram(unsigned long pagenr)
{
int i;
for (i = 0; i < boot_mem_map.nr_map; i++) {
unsigned long addr, end;
switch (boot_mem_map.map[i].type) {
case BOOT_MEM_RAM:
case BOOT_MEM_INIT_RAM:
break;
default:
/* not usable memory */
continue;
}
addr = PFN_UP(boot_mem_map.map[i].addr);
end = PFN_DOWN(boot_mem_map.map[i].addr +
boot_mem_map.map[i].size);
if (pagenr >= addr && pagenr < end)
return 1;
}
return 0;
}
void __init paging_init(void)
{
unsigned long max_zone_pfns[MAX_NR_ZONES];
unsigned long lastpfn __maybe_unused;
pagetable_init();
#ifdef CONFIG_HIGHMEM
kmap_init();
#endif
kmap_coherent_init();
#ifdef CONFIG_ZONE_DMA
max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN;
#endif
#ifdef CONFIG_ZONE_DMA32
max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN;
#endif
max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
lastpfn = max_low_pfn;
#ifdef CONFIG_HIGHMEM
max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;
lastpfn = highend_pfn;
if (cpu_has_dc_aliases && max_low_pfn != highend_pfn) {
printk(KERN_WARNING "This processor doesn't support highmem."
" %ldk highmem ignored\n",
(highend_pfn - max_low_pfn) << (PAGE_SHIFT - 10));
max_zone_pfns[ZONE_HIGHMEM] = max_low_pfn;
lastpfn = max_low_pfn;
}
#endif
free_area_init_nodes(max_zone_pfns);
}
#ifdef CONFIG_64BIT
static struct kcore_list kcore_kseg0;
#endif
void __init mem_init(void)
{
unsigned long codesize, reservedpages, datasize, initsize;
unsigned long tmp, ram;
#ifdef CONFIG_HIGHMEM
#ifdef CONFIG_DISCONTIGMEM
#error "CONFIG_HIGHMEM and CONFIG_DISCONTIGMEM dont work together yet"
#endif
max_mapnr = highend_pfn ? highend_pfn : max_low_pfn;
#else
max_mapnr = max_low_pfn;
#endif
high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT);
totalram_pages += free_all_bootmem();
totalram_pages -= setup_zero_pages(); /* Setup zeroed pages. */
reservedpages = ram = 0;
for (tmp = 0; tmp < max_low_pfn; tmp++)
if (page_is_ram(tmp) && pfn_valid(tmp)) {
ram++;
if (PageReserved(pfn_to_page(tmp)))
reservedpages++;
}
num_physpages = ram;
#ifdef CONFIG_HIGHMEM
for (tmp = highstart_pfn; tmp < highend_pfn; tmp++) {
struct page *page = pfn_to_page(tmp);
if (!page_is_ram(tmp)) {
SetPageReserved(page);
continue;
}
ClearPageReserved(page);
init_page_count(page);
__free_page(page);
totalhigh_pages++;
}
totalram_pages += totalhigh_pages;
num_physpages += totalhigh_pages;
#endif
codesize = (unsigned long) &_etext - (unsigned long) &_text;
datasize = (unsigned long) &_edata - (unsigned long) &_etext;
initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
#ifdef CONFIG_64BIT
if ((unsigned long) &_text > (unsigned long) CKSEG0)
/* The -4 is a hack so that user tools don't have to handle
the overflow. */
kclist_add(&kcore_kseg0, (void *) CKSEG0,
0x80000000 - 4, KCORE_TEXT);
#endif
printk(KERN_INFO "Memory: %luk/%luk available (%ldk kernel code, "
"%ldk reserved, %ldk data, %ldk init, %ldk highmem)\n",
nr_free_pages() << (PAGE_SHIFT-10),
ram << (PAGE_SHIFT-10),
codesize >> 10,
reservedpages << (PAGE_SHIFT-10),
datasize >> 10,
initsize >> 10,
totalhigh_pages << (PAGE_SHIFT-10));
}
#endif /* !CONFIG_NEED_MULTIPLE_NODES */
void free_init_pages(const char *what, unsigned long begin, unsigned long end)
{
unsigned long pfn;
for (pfn = PFN_UP(begin); pfn < PFN_DOWN(end); pfn++) {
struct page *page = pfn_to_page(pfn);
void *addr = phys_to_virt(PFN_PHYS(pfn));
ClearPageReserved(page);
init_page_count(page);
memset(addr, POISON_FREE_INITMEM, PAGE_SIZE);
__free_page(page);
totalram_pages++;
}
printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
}
#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
free_init_pages("initrd memory",
virt_to_phys((void *)start),
virt_to_phys((void *)end));
}
#endif
void __init_refok free_initmem(void)
{
prom_free_prom_memory();
free_init_pages("unused kernel memory",
__pa_symbol(&__init_begin),
__pa_symbol(&__init_end));
}
#ifndef CONFIG_MIPS_PGD_C0_CONTEXT
unsigned long pgd_current[NR_CPUS];
#endif
/*
* On 64-bit we've got three-level pagetables with a slightly
* different layout ...
*/
#define __page_aligned(order) __attribute__((__aligned__(PAGE_SIZE<<order)))
/*
* gcc 3.3 and older have trouble determining that PTRS_PER_PGD and PGD_ORDER
* are constants. So we use the variants from asm-offset.h until that gcc
* will officially be retired.
*/
pgd_t swapper_pg_dir[_PTRS_PER_PGD] __page_aligned(_PGD_ORDER);
#ifndef __PAGETABLE_PMD_FOLDED
pmd_t invalid_pmd_table[PTRS_PER_PMD] __page_aligned(PMD_ORDER);
#endif
pte_t invalid_pte_table[PTRS_PER_PTE] __page_aligned(PTE_ORDER);