^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0-only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * linux/arch/x86_64/mm/init.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Copyright (C) 1995 Linus Torvalds
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * Copyright (C) 2000 Pavel Machek <pavel@ucw.cz>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * Copyright (C) 2002,2003 Andi Kleen <ak@suse.de>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/signal.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/sched.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/errno.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/string.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/types.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/ptrace.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) #include <linux/mman.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) #include <linux/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/swap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/smp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/initrd.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/pagemap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <linux/memblock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) #include <linux/proc_fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <linux/pci.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <linux/pfn.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #include <linux/poison.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include <linux/dma-mapping.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #include <linux/memory.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include <linux/memory_hotplug.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #include <linux/memremap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) #include <linux/nmi.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) #include <linux/gfp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) #include <linux/kcore.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) #include <asm/processor.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) #include <asm/bios_ebda.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) #include <linux/uaccess.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) #include <asm/pgalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) #include <asm/dma.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) #include <asm/fixmap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) #include <asm/e820/api.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) #include <asm/apic.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) #include <asm/tlb.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) #include <asm/mmu_context.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) #include <asm/proto.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) #include <asm/smp.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) #include <asm/sections.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) #include <asm/kdebug.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) #include <asm/numa.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) #include <asm/set_memory.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) #include <asm/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) #include <asm/uv/uv.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) #include <asm/setup.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) #include <asm/ftrace.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) #include "mm_internal.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) #include "ident_map.c"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) #define DEFINE_POPULATE(fname, type1, type2, init) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) static inline void fname##_init(struct mm_struct *mm, \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) type1##_t *arg1, type2##_t *arg2, bool init) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) { \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) if (init) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) fname##_safe(mm, arg1, arg2); \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) else \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) fname(mm, arg1, arg2); \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) DEFINE_POPULATE(p4d_populate, p4d, pud, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) DEFINE_POPULATE(pgd_populate, pgd, p4d, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) DEFINE_POPULATE(pud_populate, pud, pmd, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) DEFINE_POPULATE(pmd_populate_kernel, pmd, pte, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) #define DEFINE_ENTRY(type1, type2, init) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) static inline void set_##type1##_init(type1##_t *arg1, \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) type2##_t arg2, bool init) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) { \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) if (init) \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) set_##type1##_safe(arg1, arg2); \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) else \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) set_##type1(arg1, arg2); \
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) DEFINE_ENTRY(p4d, p4d, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) DEFINE_ENTRY(pud, pud, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) DEFINE_ENTRY(pmd, pmd, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) DEFINE_ENTRY(pte, pte, init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) * NOTE: pagetable_init alloc all the fixmap pagetables contiguous on the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) * physical space so we can cache the place of the first one and move
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) * around without checking the pgd every time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) /* Bits supported by the hardware: */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) pteval_t __supported_pte_mask __read_mostly = ~0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) /* Bits allowed in normal kernel mappings: */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) pteval_t __default_kernel_pte_mask __read_mostly = ~0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) EXPORT_SYMBOL_GPL(__supported_pte_mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) /* Used in PAGE_KERNEL_* macros which are reasonably used out-of-tree: */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) EXPORT_SYMBOL(__default_kernel_pte_mask);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) int force_personality32;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) * noexec32=on|off
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) * Control non executable heap for 32bit processes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) * To control the stack too use noexec=off
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) * on PROT_READ does not imply PROT_EXEC for 32-bit processes (default)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) * off PROT_READ implies PROT_EXEC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) static int __init nonx32_setup(char *str)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) if (!strcmp(str, "on"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) force_personality32 &= ~READ_IMPLIES_EXEC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) else if (!strcmp(str, "off"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) force_personality32 |= READ_IMPLIES_EXEC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) __setup("noexec32=", nonx32_setup);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) static void sync_global_pgds_l5(unsigned long start, unsigned long end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) unsigned long addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) for (addr = start; addr <= end; addr = ALIGN(addr + 1, PGDIR_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) const pgd_t *pgd_ref = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135) /* Check for overflow */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) if (addr < start)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) if (pgd_none(*pgd_ref))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) spin_lock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) list_for_each_entry(page, &pgd_list, lru) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) spinlock_t *pgt_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) pgd = (pgd_t *)page_address(page) + pgd_index(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) /* the pgt_lock only for Xen */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) pgt_lock = &pgd_page_get_mm(page)->page_table_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) spin_lock(pgt_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) if (!pgd_none(*pgd_ref) && !pgd_none(*pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) BUG_ON(pgd_page_vaddr(*pgd) != pgd_page_vaddr(*pgd_ref));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) if (pgd_none(*pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) set_pgd(pgd, *pgd_ref);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) spin_unlock(pgt_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) spin_unlock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) static void sync_global_pgds_l4(unsigned long start, unsigned long end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) unsigned long addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) for (addr = start; addr <= end; addr = ALIGN(addr + 1, PGDIR_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) pgd_t *pgd_ref = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) const p4d_t *p4d_ref;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) * With folded p4d, pgd_none() is always false, we need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) * handle synchonization on p4d level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) MAYBE_BUILD_BUG_ON(pgd_none(*pgd_ref));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) p4d_ref = p4d_offset(pgd_ref, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) if (p4d_none(*p4d_ref))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) spin_lock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) list_for_each_entry(page, &pgd_list, lru) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) spinlock_t *pgt_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) pgd = (pgd_t *)page_address(page) + pgd_index(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) p4d = p4d_offset(pgd, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) /* the pgt_lock only for Xen */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) pgt_lock = &pgd_page_get_mm(page)->page_table_lock;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) spin_lock(pgt_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) if (!p4d_none(*p4d_ref) && !p4d_none(*p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) BUG_ON(p4d_page_vaddr(*p4d)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) != p4d_page_vaddr(*p4d_ref));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) if (p4d_none(*p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) set_p4d(p4d, *p4d_ref);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) spin_unlock(pgt_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) spin_unlock(&pgd_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) * When memory was added make sure all the processes MM have
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) * suitable PGD entries in the local PGD level page.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) static void sync_global_pgds(unsigned long start, unsigned long end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) if (pgtable_l5_enabled())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) sync_global_pgds_l5(start, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) sync_global_pgds_l4(start, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) * NOTE: This function is marked __ref because it calls __init function
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) * (alloc_bootmem_pages). It's safe to do it ONLY when after_bootmem == 0.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) static __ref void *spp_getpage(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) void *ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) if (after_bootmem)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) ptr = (void *) get_zeroed_page(GFP_ATOMIC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) ptr = memblock_alloc(PAGE_SIZE, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) if (!ptr || ((unsigned long)ptr & ~PAGE_MASK)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) panic("set_pte_phys: cannot allocate page data %s\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) after_bootmem ? "after bootmem" : "");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) pr_debug("spp_getpage %p\n", ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) return ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) static p4d_t *fill_p4d(pgd_t *pgd, unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) if (pgd_none(*pgd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) p4d_t *p4d = (p4d_t *)spp_getpage();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) pgd_populate(&init_mm, pgd, p4d);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) if (p4d != p4d_offset(pgd, 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) printk(KERN_ERR "PAGETABLE BUG #00! %p <-> %p\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) p4d, p4d_offset(pgd, 0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) return p4d_offset(pgd, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) static pud_t *fill_pud(p4d_t *p4d, unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257) if (p4d_none(*p4d)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) pud_t *pud = (pud_t *)spp_getpage();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) p4d_populate(&init_mm, p4d, pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) if (pud != pud_offset(p4d, 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) printk(KERN_ERR "PAGETABLE BUG #01! %p <-> %p\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) pud, pud_offset(p4d, 0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) return pud_offset(p4d, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) static pmd_t *fill_pmd(pud_t *pud, unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) if (pud_none(*pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) pmd_t *pmd = (pmd_t *) spp_getpage();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) pud_populate(&init_mm, pud, pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) if (pmd != pmd_offset(pud, 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273) printk(KERN_ERR "PAGETABLE BUG #02! %p <-> %p\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) pmd, pmd_offset(pud, 0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) return pmd_offset(pud, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) static pte_t *fill_pte(pmd_t *pmd, unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) if (pmd_none(*pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) pte_t *pte = (pte_t *) spp_getpage();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) pmd_populate_kernel(&init_mm, pmd, pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) if (pte != pte_offset_kernel(pmd, 0))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) printk(KERN_ERR "PAGETABLE BUG #03!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) return pte_offset_kernel(pmd, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) static void __set_pte_vaddr(pud_t *pud, unsigned long vaddr, pte_t new_pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) pmd_t *pmd = fill_pmd(pud, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) pte_t *pte = fill_pte(pmd, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) set_pte(pte, new_pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) * It's enough to flush this one mapping.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) * (PGE mappings get flushed as well)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) flush_tlb_one_kernel(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) void set_pte_vaddr_p4d(p4d_t *p4d_page, unsigned long vaddr, pte_t new_pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) p4d_t *p4d = p4d_page + p4d_index(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) pud_t *pud = fill_pud(p4d, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) __set_pte_vaddr(pud, vaddr, new_pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) void set_pte_vaddr_pud(pud_t *pud_page, unsigned long vaddr, pte_t new_pte)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) pud_t *pud = pud_page + pud_index(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) __set_pte_vaddr(pud, vaddr, new_pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) void set_pte_vaddr(unsigned long vaddr, pte_t pteval)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) p4d_t *p4d_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) pr_debug("set_pte_vaddr %lx to %lx\n", vaddr, native_pte_val(pteval));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) pgd = pgd_offset_k(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) if (pgd_none(*pgd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) printk(KERN_ERR
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) "PGD FIXMAP MISSING, it should be setup in head.S!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) p4d_page = p4d_offset(pgd, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) set_pte_vaddr_p4d(p4d_page, vaddr, pteval);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) pmd_t * __init populate_extra_pmd(unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) pgd = pgd_offset_k(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) p4d = fill_p4d(pgd, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) pud = fill_pud(p4d, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) return fill_pmd(pud, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) pte_t * __init populate_extra_pte(unsigned long vaddr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) pmd = populate_extra_pmd(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) return fill_pte(pmd, vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) * Create large page table mappings for a range of physical addresses.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) static void __init __init_extra_mapping(unsigned long phys, unsigned long size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) enum page_cache_mode cache)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) pgprot_t prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) pgprot_val(prot) = pgprot_val(PAGE_KERNEL_LARGE) |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) protval_4k_2_large(cachemode2protval(cache));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) BUG_ON((phys & ~PMD_MASK) || (size & ~PMD_MASK));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) for (; size; phys += PMD_SIZE, size -= PMD_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) pgd = pgd_offset_k((unsigned long)__va(phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) if (pgd_none(*pgd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) p4d = (p4d_t *) spp_getpage();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) set_pgd(pgd, __pgd(__pa(p4d) | _KERNPG_TABLE |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) _PAGE_USER));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) p4d = p4d_offset(pgd, (unsigned long)__va(phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) if (p4d_none(*p4d)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) pud = (pud_t *) spp_getpage();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) set_p4d(p4d, __p4d(__pa(pud) | _KERNPG_TABLE |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) _PAGE_USER));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) pud = pud_offset(p4d, (unsigned long)__va(phys));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) if (pud_none(*pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) pmd = (pmd_t *) spp_getpage();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) _PAGE_USER));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) pmd = pmd_offset(pud, phys);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) BUG_ON(!pmd_none(*pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) set_pmd(pmd, __pmd(phys | pgprot_val(prot)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) void __init init_extra_mapping_wb(unsigned long phys, unsigned long size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) __init_extra_mapping(phys, size, _PAGE_CACHE_MODE_WB);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) void __init init_extra_mapping_uc(unsigned long phys, unsigned long size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) __init_extra_mapping(phys, size, _PAGE_CACHE_MODE_UC);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) * The head.S code sets up the kernel high mapping:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) * from __START_KERNEL_map to __START_KERNEL_map + size (== _end-_text)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) * phys_base holds the negative offset to the kernel, which is added
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) * to the compile time generated pmds. This results in invalid pmds up
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) * to the point where we hit the physaddr 0 mapping.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) * We limit the mappings to the region from _text to _brk_end. _brk_end
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) * is rounded up to the 2MB boundary. This catches the invalid pmds as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) * well, as they are located before _text:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) void __init cleanup_highmap(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) unsigned long vaddr = __START_KERNEL_map;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) unsigned long vaddr_end = __START_KERNEL_map + KERNEL_IMAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) unsigned long end = roundup((unsigned long)_brk_end, PMD_SIZE) - 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) pmd_t *pmd = level2_kernel_pgt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) * Native path, max_pfn_mapped is not set yet.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) * Xen has valid max_pfn_mapped set in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) * arch/x86/xen/mmu.c:xen_setup_kernel_pagetable().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432) if (max_pfn_mapped)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) vaddr_end = __START_KERNEL_map + (max_pfn_mapped << PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) for (; vaddr + PMD_SIZE - 1 < vaddr_end; pmd++, vaddr += PMD_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) if (pmd_none(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) if (vaddr < (unsigned long) _text || vaddr > end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) set_pmd(pmd, __pmd(0));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) * Create PTE level page table mapping for physical addresses.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) * It returns the last physical address mapped.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) static unsigned long __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) phys_pte_init(pte_t *pte_page, unsigned long paddr, unsigned long paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) pgprot_t prot, bool init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) unsigned long pages = 0, paddr_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) unsigned long paddr_last = paddr_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) pte_t *pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) pte = pte_page + pte_index(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) i = pte_index(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) for (; i < PTRS_PER_PTE; i++, paddr = paddr_next, pte++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) paddr_next = (paddr & PAGE_MASK) + PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) if (paddr >= paddr_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) if (!after_bootmem &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) E820_TYPE_RAM) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) !e820__mapped_any(paddr & PAGE_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) E820_TYPE_RESERVED_KERN))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) set_pte_init(pte, __pte(0), init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) * We will re-use the existing mapping.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) * Xen for example has some special requirements, like mapping
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) * pagetable pages as RO. So assume someone who pre-setup
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) * these mappings are more intelligent.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) if (!pte_none(*pte)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) if (!after_bootmem)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) if (0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) pr_info(" pte=%p addr=%lx pte=%016lx\n", pte, paddr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) pfn_pte(paddr >> PAGE_SHIFT, PAGE_KERNEL).pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) set_pte_init(pte, pfn_pte(paddr >> PAGE_SHIFT, prot), init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) paddr_last = (paddr & PAGE_MASK) + PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491) update_page_count(PG_LEVEL_4K, pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) return paddr_last;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) * Create PMD level page table mapping for physical addresses. The virtual
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) * and physical address have to be aligned at this level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) * It returns the last physical address mapped.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) static unsigned long __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) phys_pmd_init(pmd_t *pmd_page, unsigned long paddr, unsigned long paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) unsigned long page_size_mask, pgprot_t prot, bool init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) unsigned long pages = 0, paddr_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) unsigned long paddr_last = paddr_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) int i = pmd_index(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) for (; i < PTRS_PER_PMD; i++, paddr = paddr_next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) pmd_t *pmd = pmd_page + pmd_index(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) pte_t *pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) pgprot_t new_prot = prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) paddr_next = (paddr & PMD_MASK) + PMD_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) if (paddr >= paddr_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) if (!after_bootmem &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) !e820__mapped_any(paddr & PMD_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) E820_TYPE_RAM) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) !e820__mapped_any(paddr & PMD_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) E820_TYPE_RESERVED_KERN))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) set_pmd_init(pmd, __pmd(0), init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) if (!pmd_none(*pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) if (!pmd_large(*pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) pte = (pte_t *)pmd_page_vaddr(*pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) paddr_last = phys_pte_init(pte, paddr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) paddr_end, prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) * If we are ok with PG_LEVEL_2M mapping, then we will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) * use the existing mapping,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) * Otherwise, we will split the large page mapping but
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) * use the same existing protection bits except for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) * large page, so that we don't violate Intel's TLB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) * Application note (317080) which says, while changing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) * the page sizes, new and old translations should
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) * not differ with respect to page frame and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) * attributes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) if (page_size_mask & (1 << PG_LEVEL_2M)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) if (!after_bootmem)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) paddr_last = paddr_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) new_prot = pte_pgprot(pte_clrhuge(*(pte_t *)pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) if (page_size_mask & (1<<PG_LEVEL_2M)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) set_pte_init((pte_t *)pmd,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) pfn_pte((paddr & PMD_MASK) >> PAGE_SHIFT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) __pgprot(pgprot_val(prot) | _PAGE_PSE)),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) paddr_last = paddr_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) pte = alloc_low_page();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) paddr_last = phys_pte_init(pte, paddr, paddr_end, new_prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) pmd_populate_kernel_init(&init_mm, pmd, pte, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) update_page_count(PG_LEVEL_2M, pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577) return paddr_last;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) * Create PUD level page table mapping for physical addresses. The virtual
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) * and physical address do not have to be aligned at this level. KASLR can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) * randomize virtual addresses up to this level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) * It returns the last physical address mapped.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) static unsigned long __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) phys_pud_init(pud_t *pud_page, unsigned long paddr, unsigned long paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) unsigned long page_size_mask, pgprot_t _prot, bool init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) unsigned long pages = 0, paddr_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) unsigned long paddr_last = paddr_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) unsigned long vaddr = (unsigned long)__va(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) int i = pud_index(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) for (; i < PTRS_PER_PUD; i++, paddr = paddr_next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) pgprot_t prot = _prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) vaddr = (unsigned long)__va(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) pud = pud_page + pud_index(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) paddr_next = (paddr & PUD_MASK) + PUD_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) if (paddr >= paddr_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) if (!after_bootmem &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) !e820__mapped_any(paddr & PUD_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) E820_TYPE_RAM) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) !e820__mapped_any(paddr & PUD_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) E820_TYPE_RESERVED_KERN))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610) set_pud_init(pud, __pud(0), init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) if (!pud_none(*pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) if (!pud_large(*pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) pmd = pmd_offset(pud, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) paddr_last = phys_pmd_init(pmd, paddr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) page_size_mask,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) * If we are ok with PG_LEVEL_1G mapping, then we will
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) * use the existing mapping.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) * Otherwise, we will split the gbpage mapping but use
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) * the same existing protection bits except for large
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) * page, so that we don't violate Intel's TLB
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) * Application note (317080) which says, while changing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) * the page sizes, new and old translations should
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) * not differ with respect to page frame and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) * attributes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) if (page_size_mask & (1 << PG_LEVEL_1G)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) if (!after_bootmem)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) paddr_last = paddr_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) prot = pte_pgprot(pte_clrhuge(*(pte_t *)pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) if (page_size_mask & (1<<PG_LEVEL_1G)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) prot = __pgprot(pgprot_val(prot) | __PAGE_KERNEL_LARGE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) set_pte_init((pte_t *)pud,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) pfn_pte((paddr & PUD_MASK) >> PAGE_SHIFT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) prot),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) paddr_last = paddr_next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) pmd = alloc_low_page();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) paddr_last = phys_pmd_init(pmd, paddr, paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) page_size_mask, prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) pud_populate_init(&init_mm, pud, pmd, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668) update_page_count(PG_LEVEL_1G, pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) return paddr_last;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) static unsigned long __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) phys_p4d_init(p4d_t *p4d_page, unsigned long paddr, unsigned long paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) unsigned long page_size_mask, pgprot_t prot, bool init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677) unsigned long vaddr, vaddr_end, vaddr_next, paddr_next, paddr_last;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) paddr_last = paddr_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) vaddr = (unsigned long)__va(paddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) vaddr_end = (unsigned long)__va(paddr_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) if (!pgtable_l5_enabled())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) return phys_pud_init((pud_t *) p4d_page, paddr, paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) page_size_mask, prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) for (; vaddr < vaddr_end; vaddr = vaddr_next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) p4d_t *p4d = p4d_page + p4d_index(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) vaddr_next = (vaddr & P4D_MASK) + P4D_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) paddr = __pa(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) if (paddr >= paddr_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) paddr_next = __pa(vaddr_next);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) if (!after_bootmem &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) !e820__mapped_any(paddr & P4D_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) E820_TYPE_RAM) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) !e820__mapped_any(paddr & P4D_MASK, paddr_next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) E820_TYPE_RESERVED_KERN))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) set_p4d_init(p4d, __p4d(0), init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) if (!p4d_none(*p4d)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) pud = pud_offset(p4d, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) paddr_last = phys_pud_init(pud, paddr, __pa(vaddr_end),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) page_size_mask, prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) pud = alloc_low_page();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) paddr_last = phys_pud_init(pud, paddr, __pa(vaddr_end),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) page_size_mask, prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) p4d_populate_init(&init_mm, p4d, pud, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) return paddr_last;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) static unsigned long __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) __kernel_physical_mapping_init(unsigned long paddr_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) unsigned long paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) unsigned long page_size_mask,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) pgprot_t prot, bool init)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) bool pgd_changed = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) unsigned long vaddr, vaddr_start, vaddr_end, vaddr_next, paddr_last;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) paddr_last = paddr_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) vaddr = (unsigned long)__va(paddr_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) vaddr_end = (unsigned long)__va(paddr_end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) vaddr_start = vaddr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) for (; vaddr < vaddr_end; vaddr = vaddr_next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) pgd_t *pgd = pgd_offset_k(vaddr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) vaddr_next = (vaddr & PGDIR_MASK) + PGDIR_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) if (pgd_val(*pgd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) p4d = (p4d_t *)pgd_page_vaddr(*pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) paddr_last = phys_p4d_init(p4d, __pa(vaddr),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) __pa(vaddr_end),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) page_size_mask,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) p4d = alloc_low_page();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) paddr_last = phys_p4d_init(p4d, __pa(vaddr), __pa(vaddr_end),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) page_size_mask, prot, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) if (pgtable_l5_enabled())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759) pgd_populate_init(&init_mm, pgd, p4d, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) p4d_populate_init(&init_mm, p4d_offset(pgd, vaddr),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) (pud_t *) p4d, init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) pgd_changed = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) if (pgd_changed)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) sync_global_pgds(vaddr_start, vaddr_end - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) return paddr_last;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) * Create page table mapping for the physical memory for specific physical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) * addresses. Note that it can only be used to populate non-present entries.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) * The virtual and physical addresses have to be aligned on PMD level
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779) * down. It returns the last physical address mapped.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) unsigned long __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) kernel_physical_mapping_init(unsigned long paddr_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) unsigned long paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) unsigned long page_size_mask, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) return __kernel_physical_mapping_init(paddr_start, paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) page_size_mask, prot, true);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) * This function is similar to kernel_physical_mapping_init() above with the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) * exception that it uses set_{pud,pmd}() instead of the set_{pud,pte}_safe()
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) * when updating the mapping. The caller is responsible to flush the TLBs after
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) * the function returns.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) unsigned long __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) kernel_physical_mapping_change(unsigned long paddr_start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) unsigned long paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799) unsigned long page_size_mask)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) return __kernel_physical_mapping_init(paddr_start, paddr_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) page_size_mask, PAGE_KERNEL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) false);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) #ifndef CONFIG_NUMA
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) void __init initmem_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) memblock_set_node(0, PHYS_ADDR_MAX, &memblock.memory, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) void __init paging_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) sparse_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) * clear the default setting with node 0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) * note: don't use nodes_clear here, that is really clearing when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) * numa support is not compiled in, and later node_set_state
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) * will not set it back.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) node_clear_state(0, N_MEMORY);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) node_clear_state(0, N_NORMAL_MEMORY);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) zone_sizes_init();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) * Memory hotplug specific functions
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) * After memory hotplug the variables max_pfn, max_low_pfn and high_memory need
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) * updating.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) static void update_end_of_memory_vars(u64 start, u64 size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) unsigned long end_pfn = PFN_UP(start + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) if (end_pfn > max_pfn) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) max_pfn = end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) max_low_pfn = end_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) int add_pages(int nid, unsigned long start_pfn, unsigned long nr_pages,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) struct mhp_params *params)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) int ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) ret = __add_pages(nid, start_pfn, nr_pages, params);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) WARN_ON_ONCE(ret);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) /* update max_pfn, max_low_pfn and high_memory */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) update_end_of_memory_vars(start_pfn << PAGE_SHIFT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) nr_pages << PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) int arch_add_memory(int nid, u64 start, u64 size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) struct mhp_params *params)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) unsigned long start_pfn = start >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) unsigned long nr_pages = size >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) init_memory_mapping(start, start + size, params->pgprot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) return add_pages(nid, start_pfn, nr_pages, params);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) #define PAGE_INUSE 0xFD
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) static void __meminit free_pagetable(struct page *page, int order)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) unsigned long magic;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879) unsigned int nr_pages = 1 << order;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) /* bootmem page has reserved flag */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) if (PageReserved(page)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) __ClearPageReserved(page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) magic = (unsigned long)page->freelist;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886) if (magic == SECTION_INFO || magic == MIX_SECTION_INFO) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) while (nr_pages--)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) put_page_bootmem(page++);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890) while (nr_pages--)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) free_reserved_page(page++);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) free_pages((unsigned long)page_address(page), order);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) static void __meminit free_hugepage_table(struct page *page,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) if (altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) vmem_altmap_free(altmap, PMD_SIZE / PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) free_pagetable(page, get_order(PMD_SIZE));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905) static void __meminit free_pte_table(pte_t *pte_start, pmd_t *pmd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) pte_t *pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) for (i = 0; i < PTRS_PER_PTE; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) pte = pte_start + i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912) if (!pte_none(*pte))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) /* free a pte talbe */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) free_pagetable(pmd_page(*pmd), 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919) pmd_clear(pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923) static void __meminit free_pmd_table(pmd_t *pmd_start, pud_t *pud)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) for (i = 0; i < PTRS_PER_PMD; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) pmd = pmd_start + i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) if (!pmd_none(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) /* free a pmd talbe */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) free_pagetable(pud_page(*pud), 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) pud_clear(pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) static void __meminit free_pud_table(pud_t *pud_start, p4d_t *p4d)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) for (i = 0; i < PTRS_PER_PUD; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) pud = pud_start + i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) if (!pud_none(*pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952) /* free a pud talbe */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) free_pagetable(p4d_page(*p4d), 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) p4d_clear(p4d);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) static void __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) remove_pte_table(pte_t *pte_start, unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) bool direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963) unsigned long next, pages = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) pte_t *pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) void *page_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) phys_addr_t phys_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) pte = pte_start + pte_index(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) for (; addr < end; addr = next, pte++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) next = (addr + PAGE_SIZE) & PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971) if (next > end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) next = end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) if (!pte_present(*pte))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978) * We mapped [0,1G) memory as identity mapping when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) * initializing, in arch/x86/kernel/head_64.S. These
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) * pagetables cannot be removed.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) phys_addr = pte_val(*pte) + (addr & PAGE_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983) if (phys_addr < (phys_addr_t)0x40000000)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) if (PAGE_ALIGNED(addr) && PAGE_ALIGNED(next)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) * Do not free direct mapping pages since they were
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) * freed when offlining, or simplely not in use.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) if (!direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) free_pagetable(pte_page(*pte), 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) pte_clear(&init_mm, addr, pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) /* For non-direct mapping, pages means nothing. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) * If we are here, we are freeing vmemmap pages since
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) * direct mapped memory ranges to be freed are aligned.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) * If we are not removing the whole page, it means
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) * other page structs in this page are being used and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) * we canot remove them. So fill the unused page_structs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) * with 0xFD, and remove the page when it is wholly
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) * filled with 0xFD.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) memset((void *)addr, PAGE_INUSE, next - addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) page_addr = page_address(pte_page(*pte));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) if (!memchr_inv(page_addr, PAGE_INUSE, PAGE_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015) free_pagetable(pte_page(*pte), 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) pte_clear(&init_mm, addr, pte);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) /* Call free_pte_table() in remove_pmd_table(). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) flush_tlb_all();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) if (direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027) update_page_count(PG_LEVEL_4K, -pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) static void __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) remove_pmd_table(pmd_t *pmd_start, unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) bool direct, struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034) unsigned long next, pages = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) pte_t *pte_base;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) void *page_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) pmd = pmd_start + pmd_index(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) for (; addr < end; addr = next, pmd++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041) next = pmd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) if (!pmd_present(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) if (pmd_large(*pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) if (IS_ALIGNED(addr, PMD_SIZE) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) IS_ALIGNED(next, PMD_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) if (!direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) free_hugepage_table(pmd_page(*pmd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) pmd_clear(pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) /* If here, we are freeing vmemmap pages. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) memset((void *)addr, PAGE_INUSE, next - addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) page_addr = page_address(pmd_page(*pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) if (!memchr_inv(page_addr, PAGE_INUSE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) PMD_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) free_hugepage_table(pmd_page(*pmd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) pmd_clear(pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1073) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1074) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1075)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) pte_base = (pte_t *)pmd_page_vaddr(*pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) remove_pte_table(pte_base, addr, next, direct);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078) free_pte_table(pte_base, pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081) /* Call free_pmd_table() in remove_pud_table(). */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) if (direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) update_page_count(PG_LEVEL_2M, -pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1084) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1085)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1086) static void __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) remove_pud_table(pud_t *pud_start, unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) struct vmem_altmap *altmap, bool direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) unsigned long next, pages = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) pmd_t *pmd_base;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) void *page_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095) pud = pud_start + pud_index(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) for (; addr < end; addr = next, pud++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) next = pud_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) if (!pud_present(*pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) if (pud_large(*pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) if (IS_ALIGNED(addr, PUD_SIZE) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) IS_ALIGNED(next, PUD_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) if (!direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) free_pagetable(pud_page(*pud),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107) get_order(PUD_SIZE));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) pud_clear(pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) pages++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114) /* If here, we are freeing vmemmap pages. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) memset((void *)addr, PAGE_INUSE, next - addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) page_addr = page_address(pud_page(*pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) if (!memchr_inv(page_addr, PAGE_INUSE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119) PUD_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) free_pagetable(pud_page(*pud),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) get_order(PUD_SIZE));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) spin_lock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) pud_clear(pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125) spin_unlock(&init_mm.page_table_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132) pmd_base = pmd_offset(pud, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) remove_pmd_table(pmd_base, addr, next, direct, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) free_pmd_table(pmd_base, pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) if (direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138) update_page_count(PG_LEVEL_1G, -pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) static void __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) remove_p4d_table(p4d_t *p4d_start, unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) struct vmem_altmap *altmap, bool direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) unsigned long next, pages = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) pud_t *pud_base;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) p4d = p4d_start + p4d_index(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) for (; addr < end; addr = next, p4d++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) next = p4d_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) if (!p4d_present(*p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156) BUILD_BUG_ON(p4d_large(*p4d));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) pud_base = pud_offset(p4d, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159) remove_pud_table(pud_base, addr, next, altmap, direct);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) * For 4-level page tables we do not want to free PUDs, but in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) * 5-level case we should free them. This code will have to change
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163) * to adapt for boot-time switching between 4 and 5 level page tables.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) if (pgtable_l5_enabled())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166) free_pud_table(pud_base, p4d);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) if (direct)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) update_page_count(PG_LEVEL_512G, -pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) /* start and end are both virtual address. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) static void __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175) remove_pagetable(unsigned long start, unsigned long end, bool direct,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) unsigned long addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) for (addr = start; addr < end; addr = next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) next = pgd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) pgd = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187) if (!pgd_present(*pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) p4d = p4d_offset(pgd, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) remove_p4d_table(p4d, addr, next, altmap, direct);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) flush_tlb_all();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197) void __ref vmemmap_free(unsigned long start, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) remove_pagetable(start, end, false, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) static void __meminit
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) kernel_physical_mapping_remove(unsigned long start, unsigned long end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) start = (unsigned long)__va(start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) end = (unsigned long)__va(end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) remove_pagetable(start, end, true, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) void __ref arch_remove_memory(int nid, u64 start, u64 size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) unsigned long start_pfn = start >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) unsigned long nr_pages = size >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218) __remove_pages(start_pfn, nr_pages, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) kernel_physical_mapping_remove(start, start + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) #endif /* CONFIG_MEMORY_HOTPLUG */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223) static struct kcore_list kcore_vsyscall;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) static void __init register_page_bootmem_info(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) #ifdef CONFIG_NUMA
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) for_each_online_node(i)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) register_page_bootmem_info_node(NODE_DATA(i));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) * Pre-allocates page-table pages for the vmalloc area in the kernel page-table.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) * Only the level which needs to be synchronized between all page-tables is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) * allocated because the synchronization can be expensive.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) static void __init preallocate_vmalloc_pages(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242) unsigned long addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) const char *lvl;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) for (addr = VMALLOC_START; addr <= VMALLOC_END; addr = ALIGN(addr + 1, PGDIR_SIZE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) pgd_t *pgd = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) lvl = "p4d";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) p4d = p4d_alloc(&init_mm, pgd, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) if (!p4d)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) goto failed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) if (pgtable_l5_enabled())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) * The goal here is to allocate all possibly required
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) * hardware page tables pointed to by the top hardware
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) * level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) * On 4-level systems, the P4D layer is folded away and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264) * the above code does no preallocation. Below, go down
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) * to the pud _software_ level to ensure the second
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) * hardware level is allocated on 4-level systems too.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) lvl = "pud";
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) pud = pud_alloc(&init_mm, p4d, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) if (!pud)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271) goto failed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) failed:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) * The pages have to be there now or they will be missing in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) * process page-tables later.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282) panic("Failed to pre-allocate %s pages for vmalloc area\n", lvl);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) void __init mem_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287) pci_iommu_alloc();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) /* clear_bss() already clear the empty_zero_page */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) /* this will put all memory onto the freelists */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) memblock_free_all();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) after_bootmem = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) x86_init.hyper.init_after_bootmem();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297) * Must be done after boot memory is put on freelist, because here we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) * might set fields in deferred struct pages that have not yet been
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299) * initialized, and memblock_free_all() initializes all the reserved
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) * deferred pages for us.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) register_page_bootmem_info();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) /* Register memory areas for /proc/kcore */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305) if (get_gate_vma(&init_mm))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) kclist_add(&kcore_vsyscall, (void *)VSYSCALL_ADDR, PAGE_SIZE, KCORE_USER);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) preallocate_vmalloc_pages();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) mem_init_print_info(NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) int __init deferred_page_init_max_threads(const struct cpumask *node_cpumask)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1315) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1316) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1317) * More CPUs always led to greater speedups on tested systems, up to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) * all the nodes' CPUs. Use all since the system is otherwise idle
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) * now.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) return max_t(int, cpumask_weight(node_cpumask), 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) int kernel_set_to_readonly;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) void mark_rodata_ro(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) unsigned long start = PFN_ALIGN(_text);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) unsigned long rodata_start = PFN_ALIGN(__start_rodata);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) unsigned long end = (unsigned long)__end_rodata_hpage_align;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) unsigned long text_end = PFN_ALIGN(_etext);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) unsigned long rodata_end = PFN_ALIGN(__end_rodata);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334) unsigned long all_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) (end - start) >> 10);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338) set_memory_ro(start, (end - start) >> PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) kernel_set_to_readonly = 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) * The rodata/data/bss/brk section (but not the kernel text!)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344) * should also be not-executable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) * We align all_end to PMD_SIZE because the existing mapping
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) * is a full PMD. If we would align _brk_end to PAGE_SIZE we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348) * split the PMD and the reminder between _brk_end and the end
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) * of the PMD will remain mapped executable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351) * Any PMD which was setup after the one which covers _brk_end
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) * has been zapped already via cleanup_highmem().
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) all_end = roundup((unsigned long)_brk_end, PMD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355) set_memory_nx(text_end, (all_end - text_end) >> PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) set_ftrace_ops_ro();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359) #ifdef CONFIG_CPA_DEBUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) printk(KERN_INFO "Testing CPA: undo %lx-%lx\n", start, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) set_memory_rw(start, (end-start) >> PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) printk(KERN_INFO "Testing CPA: again\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) set_memory_ro(start, (end-start) >> PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367) free_kernel_image_pages("unused kernel image (text/rodata gap)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368) (void *)text_end, (void *)rodata_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) free_kernel_image_pages("unused kernel image (rodata/data gap)",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) (void *)rodata_end, (void *)_sdata);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) debug_checkwx();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) int kern_addr_valid(unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377) unsigned long above = ((long)addr) >> __VIRTUAL_MASK_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) pte_t *pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) if (above != 0 && above != -1UL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387) pgd = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) if (pgd_none(*pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) p4d = p4d_offset(pgd, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392) if (!p4d_present(*p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) pud = pud_offset(p4d, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) if (!pud_present(*pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) if (pud_large(*pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) return pfn_valid(pud_pfn(*pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) pmd = pmd_offset(pud, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) if (!pmd_present(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406) if (pmd_large(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) return pfn_valid(pmd_pfn(*pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409) pte = pte_offset_kernel(pmd, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) if (pte_none(*pte))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) return pfn_valid(pte_pfn(*pte));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) * Block size is the minimum amount of memory which can be hotplugged or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) * hotremoved. It must be power of two and must be equal or larger than
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419) * MIN_MEMORY_BLOCK_SIZE.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) #define MAX_BLOCK_SIZE (2UL << 30)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423) /* Amount of ram needed to start using large blocks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) #define MEM_SIZE_FOR_LARGE_BLOCK (64UL << 30)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) /* Adjustable memory block size */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427) static unsigned long set_memory_block_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428) int __init set_memory_block_size_order(unsigned int order)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) unsigned long size = 1UL << order;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432) if (size > MEM_SIZE_FOR_LARGE_BLOCK || size < MIN_MEMORY_BLOCK_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) set_memory_block_size = size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439) static unsigned long probe_memory_block_size(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) unsigned long boot_mem_end = max_pfn << PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) unsigned long bz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) /* If memory block size has been set, then use it */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) bz = set_memory_block_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446) if (bz)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) goto done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) /* Use regular block if RAM is smaller than MEM_SIZE_FOR_LARGE_BLOCK */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) if (boot_mem_end < MEM_SIZE_FOR_LARGE_BLOCK) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) bz = MIN_MEMORY_BLOCK_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) goto done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) * Use max block size to minimize overhead on bare metal, where
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) * alignment for memory hotplug isn't a concern.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459) if (!boot_cpu_has(X86_FEATURE_HYPERVISOR)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460) bz = MAX_BLOCK_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) goto done;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) /* Find the largest allowed block size that aligns to memory end */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) for (bz = MAX_BLOCK_SIZE; bz > MIN_MEMORY_BLOCK_SIZE; bz >>= 1) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) if (IS_ALIGNED(boot_mem_end, bz))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) done:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) pr_info("x86/mm: Memory block size: %ldMB\n", bz >> 20);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) return bz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1473) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1474)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1475) static unsigned long memory_block_size_probed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) unsigned long memory_block_size_bytes(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) if (!memory_block_size_probed)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479) memory_block_size_probed = probe_memory_block_size();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481) return memory_block_size_probed;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484) #ifdef CONFIG_SPARSEMEM_VMEMMAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486) * Initialise the sparsemem vmemmap using huge-pages at the PMD level.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) static long __meminitdata addr_start, addr_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) static void __meminitdata *p_start, *p_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) static int __meminitdata node_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) static int __meminit vmemmap_populate_hugepages(unsigned long start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) unsigned long end, int node, struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) unsigned long addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) for (addr = start; addr < end; addr = next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) next = pmd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) pgd = vmemmap_pgd_populate(addr, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) if (!pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) p4d = vmemmap_p4d_populate(pgd, addr, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510) if (!p4d)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) pud = vmemmap_pud_populate(p4d, addr, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) if (!pud)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) pmd = pmd_offset(pud, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) if (pmd_none(*pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519) void *p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521) p = vmemmap_alloc_block_buf(PMD_SIZE, node, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) if (p) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) pte_t entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) entry = pfn_pte(__pa(p) >> PAGE_SHIFT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526) PAGE_KERNEL_LARGE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) set_pmd(pmd, __pmd(pte_val(entry)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529) /* check to see if we have contiguous blocks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) if (p_end != p || node_start != node) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531) if (p_start)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) pr_debug(" [%lx-%lx] PMD -> [%p-%p] on node %d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) addr_start, addr_end-1, p_start, p_end-1, node_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) addr_start = addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) node_start = node;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) p_start = p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) addr_end = addr + PMD_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) p_end = p + PMD_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) } else if (altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543) return -ENOMEM; /* no fallback */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) } else if (pmd_large(*pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545) vmemmap_verify((pte_t *)pmd, node, addr, next);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) if (vmemmap_populate_basepages(addr, next, node, NULL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1552) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1553)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1554) int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) int err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) if (end - start < PAGES_PER_SECTION * sizeof(struct page))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) err = vmemmap_populate_basepages(start, end, node, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) else if (boot_cpu_has(X86_FEATURE_PSE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562) err = vmemmap_populate_hugepages(start, end, node, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) else if (altmap) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) pr_err_once("%s: no cpu support for altmap allocations\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565) __func__);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566) err = -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) err = vmemmap_populate_basepages(start, end, node, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) if (!err)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) sync_global_pgds(start, end - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571) return err;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574) #if defined(CONFIG_MEMORY_HOTPLUG_SPARSE) && defined(CONFIG_HAVE_BOOTMEM_INFO_NODE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) void register_page_bootmem_memmap(unsigned long section_nr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) struct page *start_page, unsigned long nr_pages)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) unsigned long addr = (unsigned long)start_page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) unsigned long end = (unsigned long)(start_page + nr_pages);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581) pgd_t *pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) p4d_t *p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583) pud_t *pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) pmd_t *pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) unsigned int nr_pmd_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) struct page *page;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) for (; addr < end; addr = next) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) pte_t *pte = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) pgd = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) if (pgd_none(*pgd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593) next = (addr + PAGE_SIZE) & PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596) get_page_bootmem(section_nr, pgd_page(*pgd), MIX_SECTION_INFO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) p4d = p4d_offset(pgd, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) if (p4d_none(*p4d)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600) next = (addr + PAGE_SIZE) & PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) get_page_bootmem(section_nr, p4d_page(*p4d), MIX_SECTION_INFO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605) pud = pud_offset(p4d, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) if (pud_none(*pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) next = (addr + PAGE_SIZE) & PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) get_page_bootmem(section_nr, pud_page(*pud), MIX_SECTION_INFO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612) if (!boot_cpu_has(X86_FEATURE_PSE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) next = (addr + PAGE_SIZE) & PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) pmd = pmd_offset(pud, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615) if (pmd_none(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617) get_page_bootmem(section_nr, pmd_page(*pmd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) MIX_SECTION_INFO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) pte = pte_offset_kernel(pmd, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) if (pte_none(*pte))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) get_page_bootmem(section_nr, pte_page(*pte),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) SECTION_INFO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) next = pmd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628) pmd = pmd_offset(pud, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) if (pmd_none(*pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) nr_pmd_pages = 1 << get_order(PMD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) page = pmd_page(*pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634) while (nr_pmd_pages--)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) get_page_bootmem(section_nr, page++,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636) SECTION_INFO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) void __meminit vmemmap_populate_print_last(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) if (p_start) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) pr_debug(" [%lx-%lx] PMD -> [%p-%p] on node %d\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646) addr_start, addr_end-1, p_start, p_end-1, node_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) p_start = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648) p_end = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) node_start = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) #endif
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags