^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) * Based on arch/arm/mm/mmu.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * Copyright (C) 1995-2005 Russell King
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * Copyright (C) 2012 ARM Ltd.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) #include <linux/cache.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) #include <linux/export.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) #include <linux/kernel.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) #include <linux/errno.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13) #include <linux/init.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) #include <linux/ioport.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) #include <linux/kexec.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) #include <linux/libfdt.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/nodemask.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) #include <linux/memblock.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #include <linux/memory.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21) #include <linux/fs.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include <linux/io.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include <linux/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include <linux/vmalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <asm/barrier.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <asm/cputype.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #include <asm/fixmap.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include <asm/kasan.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #include <asm/kernel-pgtable.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include <asm/sections.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #include <asm/setup.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) #include <linux/sizes.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) #include <asm/tlb.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) #include <asm/mmu_context.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) #include <asm/ptdump.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) #include <asm/tlbflush.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) #include <asm/pgalloc.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) #define NO_BLOCK_MAPPINGS BIT(0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) #define NO_CONT_MAPPINGS BIT(1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) u64 idmap_t0sz = TCR_T0SZ(VA_BITS_MIN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) u64 idmap_ptrs_per_pgd = PTRS_PER_PGD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) u64 __section(".mmuoff.data.write") vabits_actual;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) EXPORT_SYMBOL(vabits_actual);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) u64 kimage_voffset __ro_after_init;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) EXPORT_SYMBOL(kimage_voffset);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) * Empty_zero_page is a special page that is used for zero-initialized data
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) * and COW.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) unsigned long empty_zero_page[PAGE_SIZE / sizeof(unsigned long)] __page_aligned_bss;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) EXPORT_SYMBOL(empty_zero_page);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) static pte_t bm_pte[PTRS_PER_PTE] __page_aligned_bss;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60) static pmd_t bm_pmd[PTRS_PER_PMD] __page_aligned_bss __maybe_unused;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) static pud_t bm_pud[PTRS_PER_PUD] __page_aligned_bss __maybe_unused;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) static DEFINE_SPINLOCK(swapper_pgdir_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) static DEFINE_MUTEX(fixmap_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) void set_swapper_pgd(pgd_t *pgdp, pgd_t pgd)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) pgd_t *fixmap_pgdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) spin_lock(&swapper_pgdir_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) fixmap_pgdp = pgd_set_fixmap(__pa_symbol(pgdp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) WRITE_ONCE(*fixmap_pgdp, pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) * We need dsb(ishst) here to ensure the page-table-walker sees
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) * our new entry before set_p?d() returns. The fixmap's
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) * flush_tlb_kernel_range() via clear_fixmap() does this for us.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) pgd_clear_fixmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) spin_unlock(&swapper_pgdir_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) unsigned long size, pgprot_t vma_prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) if (!pfn_valid(pfn))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) return pgprot_noncached(vma_prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) else if (file->f_flags & O_SYNC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) return pgprot_writecombine(vma_prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) return vma_prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) EXPORT_SYMBOL(phys_mem_access_prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) static phys_addr_t __init early_pgtable_alloc(int shift)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) phys_addr_t phys;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) void *ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) phys = memblock_phys_alloc(PAGE_SIZE, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) if (!phys)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) panic("Failed to allocate page table page\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) * The FIX_{PGD,PUD,PMD} slots may be in active use, but the FIX_PTE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) * slot will be free, so we can (ab)use the FIX_PTE slot to initialise
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) * any level of table.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) ptr = pte_set_fixmap(phys);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) memset(ptr, 0, PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) * Implicit barriers also ensure the zeroed page is visible to the page
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) * table walker
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) pte_clear_fixmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) return phys;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) static bool pgattr_change_is_safe(u64 old, u64 new)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) * The following mapping attributes may be updated in live
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) * kernel mappings without the need for break-before-make.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) pteval_t mask = PTE_PXN | PTE_RDONLY | PTE_WRITE | PTE_NG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) /* creating or taking down mappings is always safe */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) if (old == 0 || new == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) /* live contiguous mappings may not be manipulated at all */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) if ((old | new) & PTE_CONT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) /* Transitioning from Non-Global to Global is unsafe */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) if (old & ~new & PTE_NG)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) * Changing the memory type between Normal and Normal-Tagged is safe
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142) * since Tagged is considered a permission attribute from the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) * mismatched attribute aliases perspective.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) if (((old & PTE_ATTRINDX_MASK) == PTE_ATTRINDX(MT_NORMAL) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) (old & PTE_ATTRINDX_MASK) == PTE_ATTRINDX(MT_NORMAL_TAGGED)) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) ((new & PTE_ATTRINDX_MASK) == PTE_ATTRINDX(MT_NORMAL) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) (new & PTE_ATTRINDX_MASK) == PTE_ATTRINDX(MT_NORMAL_TAGGED)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) mask |= PTE_ATTRINDX_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) return ((old ^ new) & ~mask) == 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) static void init_pte(pmd_t *pmdp, unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) phys_addr_t phys, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) pte_t *ptep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) ptep = pte_set_fixmap_offset(pmdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) pte_t old_pte = READ_ONCE(*ptep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) set_pte(ptep, pfn_pte(__phys_to_pfn(phys), prot));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) * After the PTE entry has been populated once, we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) * only allow updates to the permission attributes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) BUG_ON(!pgattr_change_is_safe(pte_val(old_pte),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) READ_ONCE(pte_val(*ptep))));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) phys += PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) } while (ptep++, addr += PAGE_SIZE, addr != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) pte_clear_fixmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) static void alloc_init_cont_pte(pmd_t *pmdp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) unsigned long end, phys_addr_t phys,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) pgprot_t prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) phys_addr_t (*pgtable_alloc)(int),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) pmd_t pmd = READ_ONCE(*pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) BUG_ON(pmd_sect(pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) if (pmd_none(pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) phys_addr_t pte_phys;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) BUG_ON(!pgtable_alloc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191) pte_phys = pgtable_alloc(PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) __pmd_populate(pmdp, pte_phys, PMD_TYPE_TABLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) pmd = READ_ONCE(*pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) BUG_ON(pmd_bad(pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198) pgprot_t __prot = prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) next = pte_cont_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) /* use a contiguous mapping if the range is suitably aligned */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203) if ((((addr | next | phys) & ~CONT_PTE_MASK) == 0) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) (flags & NO_CONT_MAPPINGS) == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) __prot = __pgprot(pgprot_val(prot) | PTE_CONT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) init_pte(pmdp, addr, next, phys, __prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) phys += next - addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) } while (addr = next, addr != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) static void init_pmd(pud_t *pudp, unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) phys_addr_t phys, pgprot_t prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) phys_addr_t (*pgtable_alloc)(int), int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) pmd_t *pmdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) pmdp = pmd_set_fixmap_offset(pudp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) pmd_t old_pmd = READ_ONCE(*pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) next = pmd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) /* try section mapping first */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) if (((addr | next | phys) & ~SECTION_MASK) == 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) (flags & NO_BLOCK_MAPPINGS) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) pmd_set_huge(pmdp, phys, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) * After the PMD entry has been populated once, we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) * only allow updates to the permission attributes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) BUG_ON(!pgattr_change_is_safe(pmd_val(old_pmd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) READ_ONCE(pmd_val(*pmdp))));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) alloc_init_cont_pte(pmdp, addr, next, phys, prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) pgtable_alloc, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) BUG_ON(pmd_val(old_pmd) != 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) pmd_val(old_pmd) != READ_ONCE(pmd_val(*pmdp)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244) phys += next - addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) } while (pmdp++, addr = next, addr != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) pmd_clear_fixmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250) static void alloc_init_cont_pmd(pud_t *pudp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) unsigned long end, phys_addr_t phys,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) pgprot_t prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) phys_addr_t (*pgtable_alloc)(int), int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) pud_t pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) * Check for initial section mappings in the pgd/pud.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) BUG_ON(pud_sect(pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) if (pud_none(pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) phys_addr_t pmd_phys;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) BUG_ON(!pgtable_alloc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265) pmd_phys = pgtable_alloc(PMD_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) __pud_populate(pudp, pmd_phys, PUD_TYPE_TABLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) BUG_ON(pud_bad(pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) pgprot_t __prot = prot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) next = pmd_cont_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) /* use a contiguous mapping if the range is suitably aligned */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277) if ((((addr | next | phys) & ~CONT_PMD_MASK) == 0) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) (flags & NO_CONT_MAPPINGS) == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) __prot = __pgprot(pgprot_val(prot) | PTE_CONT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) init_pmd(pudp, addr, next, phys, __prot, pgtable_alloc, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) phys += next - addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) } while (addr = next, addr != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) static inline bool use_1G_block(unsigned long addr, unsigned long next,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) unsigned long phys)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) if (PAGE_SHIFT != 12)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293) if (((addr | next | phys) & ~PUD_MASK) != 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299) static void alloc_init_pud(pgd_t *pgdp, unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) phys_addr_t phys, pgprot_t prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) phys_addr_t (*pgtable_alloc)(int),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) pud_t *pudp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306) p4d_t *p4dp = p4d_offset(pgdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) p4d_t p4d = READ_ONCE(*p4dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) if (p4d_none(p4d)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310) phys_addr_t pud_phys;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) BUG_ON(!pgtable_alloc);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) pud_phys = pgtable_alloc(PUD_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) __p4d_populate(p4dp, pud_phys, PUD_TYPE_TABLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) p4d = READ_ONCE(*p4dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) BUG_ON(p4d_bad(p4d));
^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) * No need for locking during early boot. And it doesn't work as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) * expected with KASLR enabled.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) if (system_state != SYSTEM_BOOTING)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) mutex_lock(&fixmap_lock);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) pudp = pud_set_fixmap_offset(p4dp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) pud_t old_pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) next = pud_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) * For 4K granule only, attempt to put down a 1GB block
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) if (use_1G_block(addr, next, phys) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) (flags & NO_BLOCK_MAPPINGS) == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) pud_set_huge(pudp, phys, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) * After the PUD entry has been populated once, we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) * only allow updates to the permission attributes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) BUG_ON(!pgattr_change_is_safe(pud_val(old_pud),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) READ_ONCE(pud_val(*pudp))));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) alloc_init_cont_pmd(pudp, addr, next, phys, prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) pgtable_alloc, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) BUG_ON(pud_val(old_pud) != 0 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) pud_val(old_pud) != READ_ONCE(pud_val(*pudp)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) phys += next - addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) } while (pudp++, addr = next, addr != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) pud_clear_fixmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) if (system_state != SYSTEM_BOOTING)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) mutex_unlock(&fixmap_lock);
^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) static void __create_pgd_mapping(pgd_t *pgdir, phys_addr_t phys,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) unsigned long virt, phys_addr_t size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) pgprot_t prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) phys_addr_t (*pgtable_alloc)(int),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) unsigned long addr, end, next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) pgd_t *pgdp = pgd_offset_pgd(pgdir, virt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) * If the virtual and physical address don't have the same offset
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) * within a page, we cannot map the region as the caller expects.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) if (WARN_ON((phys ^ virt) & ~PAGE_MASK))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) phys &= PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) addr = virt & PAGE_MASK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) end = PAGE_ALIGN(virt + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) next = pgd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) alloc_init_pud(pgdp, addr, next, phys, prot, pgtable_alloc,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) phys += next - addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) } while (pgdp++, addr = next, addr != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) static phys_addr_t __pgd_pgtable_alloc(int shift)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) void *ptr = (void *)__get_free_page(GFP_PGTABLE_KERNEL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) BUG_ON(!ptr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) /* Ensure the zeroed page is visible to the page table walker */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) dsb(ishst);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393) return __pa(ptr);
^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) static phys_addr_t pgd_pgtable_alloc(int shift)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) phys_addr_t pa = __pgd_pgtable_alloc(shift);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) * Call proper page table ctor in case later we need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) * call core mm functions like apply_to_page_range() on
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) * this pre-allocated page table.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) * We don't select ARCH_ENABLE_SPLIT_PMD_PTLOCK if pmd is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) * folded, and if so pgtable_pmd_page_ctor() becomes nop.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) if (shift == PAGE_SHIFT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409) BUG_ON(!pgtable_pte_page_ctor(phys_to_page(pa)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) else if (shift == PMD_SHIFT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) BUG_ON(!pgtable_pmd_page_ctor(phys_to_page(pa)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) return pa;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) * This function can only be used to modify existing table entries,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418) * without allocating new levels of table. Note that this permits the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) * creation of new section or page entries.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) static void __init create_mapping_noalloc(phys_addr_t phys, unsigned long virt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422) phys_addr_t size, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424) if ((virt >= PAGE_END) && (virt < VMALLOC_START)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) pr_warn("BUG: not creating mapping for %pa at 0x%016lx - outside kernel range\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) &phys, virt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) __create_pgd_mapping(init_mm.pgd, phys, virt, size, prot, NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) NO_CONT_MAPPINGS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 431) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 432)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 433) void __init create_pgd_mapping(struct mm_struct *mm, phys_addr_t phys,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) unsigned long virt, phys_addr_t size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) pgprot_t prot, bool page_mappings_only)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) int flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) BUG_ON(mm == &init_mm);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) if (page_mappings_only)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) flags = NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) __create_pgd_mapping(mm->pgd, phys, virt, size, prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) pgd_pgtable_alloc, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) static void update_mapping_prot(phys_addr_t phys, unsigned long virt,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) phys_addr_t size, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) if ((virt >= PAGE_END) && (virt < VMALLOC_START)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452) pr_warn("BUG: not updating mapping for %pa at 0x%016lx - outside kernel range\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) &phys, virt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457) __create_pgd_mapping(init_mm.pgd, phys, virt, size, prot, NULL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) NO_CONT_MAPPINGS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460) /* flush the TLBs after updating live kernel mappings */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) flush_tlb_kernel_range(virt, virt + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) static void __init __map_memblock(pgd_t *pgdp, phys_addr_t start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) phys_addr_t end, pgprot_t prot, int flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) __create_pgd_mapping(pgdp, start, __phys_to_virt(start), end - start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) prot, early_pgtable_alloc, flags);
^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) void __init mark_linear_text_alias_ro(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) * Remove the write permissions from the linear alias of .text/.rodata
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) update_mapping_prot(__pa_symbol(_text), (unsigned long)lm_alias(_text),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) (unsigned long)__init_begin - (unsigned long)_text,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478) PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481) static bool crash_mem_map __initdata;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) static int __init enable_crash_mem_map(char *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) * Proper parameter parsing is done by reserve_crashkernel(). We only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) * need to know if the linear map has to avoid block mappings so that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) * the crashkernel reservations can be unmapped later.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) crash_mem_map = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) early_param("crashkernel", enable_crash_mem_map);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) static void __init map_mem(pgd_t *pgdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) phys_addr_t kernel_start = __pa_symbol(_text);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499) phys_addr_t kernel_end = __pa_symbol(__init_begin);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) phys_addr_t start, end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) int flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502) u64 i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) if (rodata_full || debug_pagealloc_enabled() ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505) IS_ENABLED(CONFIG_KFENCE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) flags = NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509) * Take care not to create a writable alias for the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) * read-only text and rodata sections of the kernel image.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) * So temporarily mark them as NOMAP to skip mappings in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) * the following for-loop
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) memblock_mark_nomap(kernel_start, kernel_end - kernel_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) #ifdef CONFIG_KEXEC_CORE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) if (crash_mem_map) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) if (IS_ENABLED(CONFIG_ZONE_DMA) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) IS_ENABLED(CONFIG_ZONE_DMA32))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520) flags |= NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) else if (crashk_res.end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) memblock_mark_nomap(crashk_res.start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) resource_size(&crashk_res));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) /* map all the memory banks */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) for_each_mem_range(i, &start, &end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529) if (start >= end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) * The linear map must allow allocation tags reading/writing
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) * if MTE is present. Otherwise, it has the same attributes as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) * PAGE_KERNEL.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) __map_memblock(pgdp, start, end, pgprot_tagged(PAGE_KERNEL),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) * Map the linear alias of the [_text, __init_begin) interval
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542) * as non-executable now, and remove the write permission in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) * mark_linear_text_alias_ro() below (which will be called after
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) * alternative patching has completed). This makes the contents
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) * of the region accessible to subsystems such as hibernate,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) * but protects it from inadvertent modification or execution.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) * Note that contiguous mappings cannot be remapped in this way,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) * so we should avoid them here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) __map_memblock(pgdp, kernel_start, kernel_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551) PAGE_KERNEL, NO_CONT_MAPPINGS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) memblock_clear_nomap(kernel_start, kernel_end - kernel_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) * Use page-level mappings here so that we can shrink the region
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) * in page granularity and put back unused memory to buddy system
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) * through /sys/kernel/kexec_crash_size interface.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559) #ifdef CONFIG_KEXEC_CORE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) if (crash_mem_map &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) !IS_ENABLED(CONFIG_ZONE_DMA) && !IS_ENABLED(CONFIG_ZONE_DMA32)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) if (crashk_res.end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) __map_memblock(pgdp, crashk_res.start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564) crashk_res.end + 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) PAGE_KERNEL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) memblock_clear_nomap(crashk_res.start,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568) resource_size(&crashk_res));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) void mark_rodata_ro(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) unsigned long section_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 577)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 578) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 579) * mark .rodata as read only. Use __init_begin rather than __end_rodata
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) * to cover NOTES and EXCEPTION_TABLE.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) section_size = (unsigned long)__init_begin - (unsigned long)__start_rodata;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583) update_mapping_prot(__pa_symbol(__start_rodata), (unsigned long)__start_rodata,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) section_size, PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) debug_checkwx();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) static void __init map_kernel_segment(pgd_t *pgdp, void *va_start, void *va_end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590) pgprot_t prot, struct vm_struct *vma,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) int flags, unsigned long vm_flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) phys_addr_t pa_start = __pa_symbol(va_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594) unsigned long size = va_end - va_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) BUG_ON(!PAGE_ALIGNED(pa_start));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) BUG_ON(!PAGE_ALIGNED(size));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599) __create_pgd_mapping(pgdp, pa_start, (unsigned long)va_start, size, prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) early_pgtable_alloc, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) if (!(vm_flags & VM_NO_GUARD))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603) size += PAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) vma->addr = va_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606) vma->phys_addr = pa_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) vma->size = size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) vma->flags = VM_MAP | vm_flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) vma->caller = __builtin_return_address(0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) vm_area_add_early(vma);
^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) static int __init parse_rodata(char *arg)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616) int ret = strtobool(arg, &rodata_enabled);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) if (!ret) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) rodata_full = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622) /* permit 'full' in addition to boolean options */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) if (strcmp(arg, "full"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) rodata_enabled = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627) rodata_full = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) early_param("rodata", parse_rodata);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) #ifdef CONFIG_UNMAP_KERNEL_AT_EL0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) static int __init map_entry_trampoline(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) pgprot_t prot = rodata_enabled ? PAGE_KERNEL_ROX : PAGE_KERNEL_EXEC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) phys_addr_t pa_start = __pa_symbol(__entry_tramp_text_start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) /* The trampoline is always mapped and can therefore be global */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) pgprot_val(prot) &= ~PTE_NG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) /* Map only the text into the trampoline page table */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) memset(tramp_pg_dir, 0, PGD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) __create_pgd_mapping(tramp_pg_dir, pa_start, TRAMP_VALIAS,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646) entry_tramp_text_size(), prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) __pgd_pgtable_alloc, NO_BLOCK_MAPPINGS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) /* Map both the text and data into the kernel page table */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) for (i = 0; i < DIV_ROUND_UP(entry_tramp_text_size(), PAGE_SIZE); i++)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) __set_fixmap(FIX_ENTRY_TRAMP_TEXT1 - i,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) pa_start + i * PAGE_SIZE, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) if (IS_ENABLED(CONFIG_RANDOMIZE_BASE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655) extern char __entry_tramp_data_start[];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) __set_fixmap(FIX_ENTRY_TRAMP_DATA,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658) __pa_symbol(__entry_tramp_data_start),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) PAGE_KERNEL_RO);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) core_initcall(map_entry_trampoline);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) #endif
^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) * Open coded check for BTI, only for use to determine configuration
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) * for early mappings for before the cpufeature code has run.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) static bool arm64_early_this_cpu_has_bti(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) u64 pfr1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) if (!IS_ENABLED(CONFIG_ARM64_BTI_KERNEL))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) pfr1 = __read_sysreg_by_encoding(SYS_ID_AA64PFR1_EL1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) return cpuid_feature_extract_unsigned_field(pfr1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) ID_AA64PFR1_BT_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) * Create fine-grained mappings for the kernel.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) static void __init map_kernel(pgd_t *pgdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) static struct vm_struct vmlinux_text, vmlinux_rodata, vmlinux_inittext,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) vmlinux_initdata, vmlinux_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) * External debuggers may need to write directly to the text
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) * mapping to install SW breakpoints. Allow this (only) when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694) * explicitly requested with rodata=off.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) pgprot_t text_prot = rodata_enabled ? PAGE_KERNEL_ROX : PAGE_KERNEL_EXEC;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699) * If we have a CPU that supports BTI and a kernel built for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) * BTI then mark the kernel executable text as guarded pages
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) * now so we don't have to rewrite the page tables later.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) if (arm64_early_this_cpu_has_bti())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) text_prot = __pgprot_modify(text_prot, PTE_GP, PTE_GP);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) * Only rodata will be remapped with different permissions later on,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) * all other segments are allowed to use contiguous mappings.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710) map_kernel_segment(pgdp, _text, _etext, text_prot, &vmlinux_text, 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) VM_NO_GUARD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) map_kernel_segment(pgdp, __start_rodata, __inittext_begin, PAGE_KERNEL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) &vmlinux_rodata, NO_CONT_MAPPINGS, VM_NO_GUARD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) map_kernel_segment(pgdp, __inittext_begin, __inittext_end, text_prot,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) &vmlinux_inittext, 0, VM_NO_GUARD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) map_kernel_segment(pgdp, __initdata_begin, __initdata_end, PAGE_KERNEL,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) &vmlinux_initdata, 0, VM_NO_GUARD);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) map_kernel_segment(pgdp, _data, _end, PAGE_KERNEL, &vmlinux_data, 0, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) if (!READ_ONCE(pgd_val(*pgd_offset_pgd(pgdp, FIXADDR_START)))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722) * The fixmap falls in a separate pgd to the kernel, and doesn't
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) * live in the carveout for the swapper_pg_dir. We can simply
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) * re-use the existing dir for the fixmap.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) set_pgd(pgd_offset_pgd(pgdp, FIXADDR_START),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) READ_ONCE(*pgd_offset_k(FIXADDR_START)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) } else if (CONFIG_PGTABLE_LEVELS > 3) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) pgd_t *bm_pgdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) p4d_t *bm_p4dp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) pud_t *bm_pudp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) * The fixmap shares its top level pgd entry with the kernel
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) * mapping. This can really only occur when we are running
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) * with 16k/4 levels, so we can simply reuse the pud level
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736) * entry instead.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) BUG_ON(!IS_ENABLED(CONFIG_ARM64_16K_PAGES));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) bm_pgdp = pgd_offset_pgd(pgdp, FIXADDR_START);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740) bm_p4dp = p4d_offset(bm_pgdp, FIXADDR_START);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) bm_pudp = pud_set_fixmap_offset(bm_p4dp, FIXADDR_START);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) pud_populate(&init_mm, bm_pudp, lm_alias(bm_pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) pud_clear_fixmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) BUG();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) kasan_copy_shadow(pgdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) void __init paging_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) pgd_t *pgdp = pgd_set_fixmap(__pa_symbol(swapper_pg_dir));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) map_kernel(pgdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) map_mem(pgdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) pgd_clear_fixmap();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) cpu_replace_ttbr1(lm_alias(swapper_pg_dir));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) init_mm.pgd = swapper_pg_dir;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) memblock_free(__pa_symbol(init_pg_dir),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) __pa_symbol(init_pg_end) - __pa_symbol(init_pg_dir));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766) memblock_allow_resize();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) * Check whether a kernel address is valid (derived from arch/x86/).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) int kern_addr_valid(unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) pgd_t *pgdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) p4d_t *p4dp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) pud_t *pudp, pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) pmd_t *pmdp, pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) pte_t *ptep, pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) addr = arch_kasan_reset_tag(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) if ((((long)addr) >> VA_BITS) != -1UL)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784) pgdp = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) if (pgd_none(READ_ONCE(*pgdp)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788) p4dp = p4d_offset(pgdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) if (p4d_none(READ_ONCE(*p4dp)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) pudp = pud_offset(p4dp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794) if (pud_none(pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797) if (pud_sect(pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) return pfn_valid(pud_pfn(pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) pmdp = pmd_offset(pudp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) pmd = READ_ONCE(*pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) if (pmd_none(pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 803) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 804)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 805) if (pmd_sect(pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) return pfn_valid(pmd_pfn(pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808) ptep = pte_offset_kernel(pmdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) pte = READ_ONCE(*ptep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) if (pte_none(pte))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813) return pfn_valid(pte_pfn(pte));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) static void free_hotplug_page_range(struct page *page, size_t size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820) if (altmap) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) vmem_altmap_free(altmap, size >> PAGE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) WARN_ON(PageReserved(page));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) free_pages((unsigned long)page_address(page), get_order(size));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 825) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 826) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 827)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) static void free_hotplug_pgtable_page(struct page *page)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) free_hotplug_page_range(page, PAGE_SIZE, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) static bool pgtable_range_aligned(unsigned long start, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) unsigned long floor, unsigned long ceiling,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) unsigned long mask)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837) start &= mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) if (start < floor)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) if (ceiling) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842) ceiling &= mask;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) if (!ceiling)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844) return false;
^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) if (end - 1 > ceiling - 1)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852) static void unmap_hotplug_pte_range(pmd_t *pmdp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) unsigned long end, bool free_mapped,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) pte_t *ptep, pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) ptep = pte_offset_kernel(pmdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) pte = READ_ONCE(*ptep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861) if (pte_none(pte))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) WARN_ON(!pte_present(pte));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) pte_clear(&init_mm, addr, ptep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) if (free_mapped)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) free_hotplug_page_range(pte_page(pte),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) PAGE_SIZE, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) } while (addr += PAGE_SIZE, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 871) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 872)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 873) static void unmap_hotplug_pmd_range(pud_t *pudp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) unsigned long end, bool free_mapped,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 878) pmd_t *pmdp, pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 879)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 880) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 881) next = pmd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 882) pmdp = pmd_offset(pudp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 883) pmd = READ_ONCE(*pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 884) if (pmd_none(pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 885) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 886)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 887) WARN_ON(!pmd_present(pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 888) if (pmd_sect(pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 889) pmd_clear(pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 890)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 891) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 892) * One TLBI should be sufficient here as the PMD_SIZE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 893) * range is mapped with a single block entry.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 894) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 895) flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 896) if (free_mapped)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 897) free_hotplug_page_range(pmd_page(pmd),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 898) PMD_SIZE, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 899) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 900) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 901) WARN_ON(!pmd_table(pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 902) unmap_hotplug_pte_range(pmdp, addr, next, free_mapped, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 903) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 904) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 905)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 906) static void unmap_hotplug_pud_range(p4d_t *p4dp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 907) unsigned long end, bool free_mapped,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 908) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 909) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 910) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 911) pud_t *pudp, pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 912)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 913) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 914) next = pud_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 915) pudp = pud_offset(p4dp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 916) pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 917) if (pud_none(pud))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 918) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 919)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 920) WARN_ON(!pud_present(pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 921) if (pud_sect(pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 922) pud_clear(pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 923)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 924) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 925) * One TLBI should be sufficient here as the PUD_SIZE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 926) * range is mapped with a single block entry.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 927) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 928) flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 929) if (free_mapped)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 930) free_hotplug_page_range(pud_page(pud),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 931) PUD_SIZE, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 932) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 933) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 934) WARN_ON(!pud_table(pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 935) unmap_hotplug_pmd_range(pudp, addr, next, free_mapped, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 936) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 937) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 938)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 939) static void unmap_hotplug_p4d_range(pgd_t *pgdp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 940) unsigned long end, bool free_mapped,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 941) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 942) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 943) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 944) p4d_t *p4dp, p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 945)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 946) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 947) next = p4d_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 948) p4dp = p4d_offset(pgdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 949) p4d = READ_ONCE(*p4dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 950) if (p4d_none(p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 951) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 952)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 953) WARN_ON(!p4d_present(p4d));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 954) unmap_hotplug_pud_range(p4dp, addr, next, free_mapped, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 955) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 956) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 957)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 958) static void unmap_hotplug_range(unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 959) bool free_mapped, struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 960) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 961) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 962) pgd_t *pgdp, pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 963)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 964) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 965) * altmap can only be used as vmemmap mapping backing memory.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 966) * In case the backing memory itself is not being freed, then
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 967) * altmap is irrelevant. Warn about this inconsistency when
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 968) * encountered.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 969) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 970) WARN_ON(!free_mapped && altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 971)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 972) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 973) next = pgd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 974) pgdp = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 975) pgd = READ_ONCE(*pgdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 976) if (pgd_none(pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 977) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 978)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 979) WARN_ON(!pgd_present(pgd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 980) unmap_hotplug_p4d_range(pgdp, addr, next, free_mapped, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 981) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 982) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 983)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 984) static void free_empty_pte_table(pmd_t *pmdp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 985) unsigned long end, unsigned long floor,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 986) unsigned long ceiling)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 987) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 988) pte_t *ptep, pte;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 989) unsigned long i, start = addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 990)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 991) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 992) ptep = pte_offset_kernel(pmdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 993) pte = READ_ONCE(*ptep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 994)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 995) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 996) * This is just a sanity check here which verifies that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 997) * pte clearing has been done by earlier unmap loops.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 998) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 999) WARN_ON(!pte_none(pte));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1000) } while (addr += PAGE_SIZE, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1001)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1002) if (!pgtable_range_aligned(start, end, floor, ceiling, PMD_MASK))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1003) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1004)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1005) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1006) * Check whether we can free the pte page if the rest of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1007) * entries are empty. Overlap with other regions have been
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1008) * handled by the floor/ceiling check.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1009) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1010) ptep = pte_offset_kernel(pmdp, 0UL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1011) for (i = 0; i < PTRS_PER_PTE; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1012) if (!pte_none(READ_ONCE(ptep[i])))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1013) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1014) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1015)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1016) pmd_clear(pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1017) __flush_tlb_kernel_pgtable(start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1018) free_hotplug_pgtable_page(virt_to_page(ptep));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1019) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1020)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1021) static void free_empty_pmd_table(pud_t *pudp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1022) unsigned long end, unsigned long floor,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1023) unsigned long ceiling)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1024) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1025) pmd_t *pmdp, pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1026) unsigned long i, next, start = addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1027)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1028) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1029) next = pmd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1030) pmdp = pmd_offset(pudp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1031) pmd = READ_ONCE(*pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1032) if (pmd_none(pmd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1033) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1034)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1035) WARN_ON(!pmd_present(pmd) || !pmd_table(pmd) || pmd_sect(pmd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1036) free_empty_pte_table(pmdp, addr, next, floor, ceiling);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1037) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1038)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1039) if (CONFIG_PGTABLE_LEVELS <= 2)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1040) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1041)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1042) if (!pgtable_range_aligned(start, end, floor, ceiling, PUD_MASK))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1043) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1044)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1045) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1046) * Check whether we can free the pmd page if the rest of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1047) * entries are empty. Overlap with other regions have been
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1048) * handled by the floor/ceiling check.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1049) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1050) pmdp = pmd_offset(pudp, 0UL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1051) for (i = 0; i < PTRS_PER_PMD; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1052) if (!pmd_none(READ_ONCE(pmdp[i])))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1053) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1054) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1055)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1056) pud_clear(pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1057) __flush_tlb_kernel_pgtable(start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1058) free_hotplug_pgtable_page(virt_to_page(pmdp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1059) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1060)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1061) static void free_empty_pud_table(p4d_t *p4dp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1062) unsigned long end, unsigned long floor,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1063) unsigned long ceiling)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1064) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1065) pud_t *pudp, pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1066) unsigned long i, next, start = addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1067)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1068) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1069) next = pud_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1070) pudp = pud_offset(p4dp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1071) pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1072) if (pud_none(pud))
^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) WARN_ON(!pud_present(pud) || !pud_table(pud) || pud_sect(pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1076) free_empty_pmd_table(pudp, addr, next, floor, ceiling);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1077) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1078)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1079) if (CONFIG_PGTABLE_LEVELS <= 3)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1080) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1081)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1082) if (!pgtable_range_aligned(start, end, floor, ceiling, PGDIR_MASK))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1083) return;
^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) * Check whether we can free the pud page if the rest of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1087) * entries are empty. Overlap with other regions have been
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1088) * handled by the floor/ceiling check.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1089) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1090) pudp = pud_offset(p4dp, 0UL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1091) for (i = 0; i < PTRS_PER_PUD; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1092) if (!pud_none(READ_ONCE(pudp[i])))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1093) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1094) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1095)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1096) p4d_clear(p4dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1097) __flush_tlb_kernel_pgtable(start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1098) free_hotplug_pgtable_page(virt_to_page(pudp));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1099) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1100)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1101) static void free_empty_p4d_table(pgd_t *pgdp, unsigned long addr,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1102) unsigned long end, unsigned long floor,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1103) unsigned long ceiling)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1104) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1105) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1106) p4d_t *p4dp, p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1108) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1109) next = p4d_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1110) p4dp = p4d_offset(pgdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1111) p4d = READ_ONCE(*p4dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1112) if (p4d_none(p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1113) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1114)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1115) WARN_ON(!p4d_present(p4d));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1116) free_empty_pud_table(p4dp, addr, next, floor, ceiling);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1117) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1118) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1119)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1120) static void free_empty_tables(unsigned long addr, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1121) unsigned long floor, unsigned long ceiling)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1122) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1123) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1124) pgd_t *pgdp, pgd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1126) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1127) next = pgd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1128) pgdp = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1129) pgd = READ_ONCE(*pgdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1130) if (pgd_none(pgd))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1131) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1132)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1133) WARN_ON(!pgd_present(pgd));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1134) free_empty_p4d_table(pgdp, addr, next, floor, ceiling);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1135) } while (addr = next, addr < end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1136) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1137) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1139) #ifdef CONFIG_SPARSEMEM_VMEMMAP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1140) #if !ARM64_SWAPPER_USES_SECTION_MAPS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1141) int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1142) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1143) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1144) return vmemmap_populate_basepages(start, end, node, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1145) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1146) #else /* !ARM64_SWAPPER_USES_SECTION_MAPS */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1147) int __meminit vmemmap_populate(unsigned long start, unsigned long end, int node,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1148) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1149) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1150) unsigned long addr = start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1151) unsigned long next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1152) pgd_t *pgdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1153) p4d_t *p4dp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1154) pud_t *pudp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1155) pmd_t *pmdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1156)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1157) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1158) next = pmd_addr_end(addr, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1159)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1160) pgdp = vmemmap_pgd_populate(addr, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1161) if (!pgdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1162) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1164) p4dp = vmemmap_p4d_populate(pgdp, addr, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1165) if (!p4dp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1166) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1167)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1168) pudp = vmemmap_pud_populate(p4dp, addr, node);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1169) if (!pudp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1170) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1171)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1172) pmdp = pmd_offset(pudp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1173) if (pmd_none(READ_ONCE(*pmdp))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1174) void *p = NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1175)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1176) p = vmemmap_alloc_block_buf(PMD_SIZE, node, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1177) if (!p) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1178) if (vmemmap_populate_basepages(addr, next, node, altmap))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1179) return -ENOMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1180) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1181) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1182)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1183) pmd_set_huge(pmdp, __pa(p), __pgprot(PROT_SECT_NORMAL));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1184) } else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1185) vmemmap_verify((pte_t *)pmdp, node, addr, next);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1186) } while (addr = next, addr != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1188) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1189) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1190) #endif /* !ARM64_SWAPPER_USES_SECTION_MAPS */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1191) void vmemmap_free(unsigned long start, unsigned long end,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1192) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1193) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1194) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1195) WARN_ON((start < VMEMMAP_START) || (end > VMEMMAP_END));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1196)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1197) unmap_hotplug_range(start, end, true, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1198) free_empty_tables(start, end, VMEMMAP_START, VMEMMAP_END);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1199) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1200) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1201) #endif /* CONFIG_SPARSEMEM_VMEMMAP */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1202)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1203) static inline pud_t * fixmap_pud(unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1204) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1205) pgd_t *pgdp = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1206) p4d_t *p4dp = p4d_offset(pgdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1207) p4d_t p4d = READ_ONCE(*p4dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1208)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1209) BUG_ON(p4d_none(p4d) || p4d_bad(p4d));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1210)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1211) return pud_offset_kimg(p4dp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1212) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1214) static inline pmd_t * fixmap_pmd(unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1215) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1216) pud_t *pudp = fixmap_pud(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1217) pud_t pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1218)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1219) BUG_ON(pud_none(pud) || pud_bad(pud));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1220)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1221) return pmd_offset_kimg(pudp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1222) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1223)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1224) static inline pte_t * fixmap_pte(unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1225) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1226) return &bm_pte[pte_index(addr)];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1227) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1228)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1229) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1230) * The p*d_populate functions call virt_to_phys implicitly so they can't be used
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1231) * directly on kernel symbols (bm_p*d). This function is called too early to use
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1232) * lm_alias so __p*d_populate functions must be used to populate with the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1233) * physical address from __pa_symbol.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1234) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1235) void __init early_fixmap_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1236) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1237) pgd_t *pgdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1238) p4d_t *p4dp, p4d;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1239) pud_t *pudp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1240) pmd_t *pmdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1241) unsigned long addr = FIXADDR_START;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1242)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1243) pgdp = pgd_offset_k(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1244) p4dp = p4d_offset(pgdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1245) p4d = READ_ONCE(*p4dp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1246) if (CONFIG_PGTABLE_LEVELS > 3 &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1247) !(p4d_none(p4d) || p4d_page_paddr(p4d) == __pa_symbol(bm_pud))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1248) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1249) * We only end up here if the kernel mapping and the fixmap
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1250) * share the top level pgd entry, which should only happen on
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1251) * 16k/4 levels configurations.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1252) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1253) BUG_ON(!IS_ENABLED(CONFIG_ARM64_16K_PAGES));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1254) pudp = pud_offset_kimg(p4dp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1255) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1256) if (p4d_none(p4d))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1257) __p4d_populate(p4dp, __pa_symbol(bm_pud), PUD_TYPE_TABLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1258) pudp = fixmap_pud(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1259) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1260) if (pud_none(READ_ONCE(*pudp)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1261) __pud_populate(pudp, __pa_symbol(bm_pmd), PMD_TYPE_TABLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1262) pmdp = fixmap_pmd(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1263) __pmd_populate(pmdp, __pa_symbol(bm_pte), PMD_TYPE_TABLE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1264)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1265) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1266) * The boot-ioremap range spans multiple pmds, for which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1267) * we are not prepared:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1268) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1269) BUILD_BUG_ON((__fix_to_virt(FIX_BTMAP_BEGIN) >> PMD_SHIFT)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1270) != (__fix_to_virt(FIX_BTMAP_END) >> PMD_SHIFT));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1271)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1272) if ((pmdp != fixmap_pmd(fix_to_virt(FIX_BTMAP_BEGIN)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1273) || pmdp != fixmap_pmd(fix_to_virt(FIX_BTMAP_END))) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1274) WARN_ON(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1275) pr_warn("pmdp %p != %p, %p\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1276) pmdp, fixmap_pmd(fix_to_virt(FIX_BTMAP_BEGIN)),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1277) fixmap_pmd(fix_to_virt(FIX_BTMAP_END)));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1278) pr_warn("fix_to_virt(FIX_BTMAP_BEGIN): %08lx\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1279) fix_to_virt(FIX_BTMAP_BEGIN));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1280) pr_warn("fix_to_virt(FIX_BTMAP_END): %08lx\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1281) fix_to_virt(FIX_BTMAP_END));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1282)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1283) pr_warn("FIX_BTMAP_END: %d\n", FIX_BTMAP_END);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1284) pr_warn("FIX_BTMAP_BEGIN: %d\n", FIX_BTMAP_BEGIN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1285) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1286) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1287)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1288) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1289) * Unusually, this is also called in IRQ context (ghes_iounmap_irq) so if we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1290) * ever need to use IPIs for TLB broadcasting, then we're in trouble here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1291) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1292) void __set_fixmap(enum fixed_addresses idx,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1293) phys_addr_t phys, pgprot_t flags)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1294) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1295) unsigned long addr = __fix_to_virt(idx);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1296) pte_t *ptep;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1297)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1298) BUG_ON(idx <= FIX_HOLE || idx >= __end_of_fixed_addresses);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1299)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1300) ptep = fixmap_pte(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1301)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1302) if (pgprot_val(flags)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1303) set_pte(ptep, pfn_pte(phys >> PAGE_SHIFT, flags));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1304) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1305) pte_clear(&init_mm, addr, ptep);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1306) flush_tlb_kernel_range(addr, addr+PAGE_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1307) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1308) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1309)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1310) void *__init fixmap_remap_fdt(phys_addr_t dt_phys, int *size, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1311) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1312) const u64 dt_virt_base = __fix_to_virt(FIX_FDT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1313) int offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1314) void *dt_virt;
^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) * Check whether the physical FDT address is set and meets the minimum
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1318) * alignment requirement. Since we are relying on MIN_FDT_ALIGN to be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1319) * at least 8 bytes so that we can always access the magic and size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1320) * fields of the FDT header after mapping the first chunk, double check
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1321) * here if that is indeed the case.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1322) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1323) BUILD_BUG_ON(MIN_FDT_ALIGN < 8);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1324) if (!dt_phys || dt_phys % MIN_FDT_ALIGN)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1325) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1326)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1327) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1328) * Make sure that the FDT region can be mapped without the need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1329) * allocate additional translation table pages, so that it is safe
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1330) * to call create_mapping_noalloc() this early.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1331) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1332) * On 64k pages, the FDT will be mapped using PTEs, so we need to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1333) * be in the same PMD as the rest of the fixmap.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1334) * On 4k pages, we'll use section mappings for the FDT so we only
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1335) * have to be in the same PUD.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1336) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1337) BUILD_BUG_ON(dt_virt_base % SZ_2M);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1338)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1339) BUILD_BUG_ON(__fix_to_virt(FIX_FDT_END) >> SWAPPER_TABLE_SHIFT !=
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1340) __fix_to_virt(FIX_BTMAP_BEGIN) >> SWAPPER_TABLE_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1341)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1342) offset = dt_phys % SWAPPER_BLOCK_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1343) dt_virt = (void *)dt_virt_base + offset;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1344)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1345) /* map the first chunk so we can read the size from the header */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1346) create_mapping_noalloc(round_down(dt_phys, SWAPPER_BLOCK_SIZE),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1347) dt_virt_base, SWAPPER_BLOCK_SIZE, prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1348)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1349) if (fdt_magic(dt_virt) != FDT_MAGIC)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1350) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1351)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1352) *size = fdt_totalsize(dt_virt);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1353) if (*size > MAX_FDT_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1354) return NULL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1355)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1356) if (offset + *size > SWAPPER_BLOCK_SIZE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1357) create_mapping_noalloc(round_down(dt_phys, SWAPPER_BLOCK_SIZE), dt_virt_base,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1358) round_up(offset + *size, SWAPPER_BLOCK_SIZE), prot);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1359)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1360) return dt_virt;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1361) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1362)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1363) int __init arch_ioremap_p4d_supported(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1364) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1365) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1366) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1367)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1368) int __init arch_ioremap_pud_supported(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1369) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1370) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1371) * Only 4k granule supports level 1 block mappings.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1372) * SW table walks can't handle removal of intermediate entries.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1373) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1374) return IS_ENABLED(CONFIG_ARM64_4K_PAGES) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1375) !IS_ENABLED(CONFIG_PTDUMP_DEBUGFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1376) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1377)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1378) int __init arch_ioremap_pmd_supported(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1379) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1380) /* See arch_ioremap_pud_supported() */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1381) return !IS_ENABLED(CONFIG_PTDUMP_DEBUGFS);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1382) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1383)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1384) int pud_set_huge(pud_t *pudp, phys_addr_t phys, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1385) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1386) pud_t new_pud = pfn_pud(__phys_to_pfn(phys), mk_pud_sect_prot(prot));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1387)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1388) /* Only allow permission changes for now */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1389) if (!pgattr_change_is_safe(READ_ONCE(pud_val(*pudp)),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1390) pud_val(new_pud)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1391) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1392)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1393) VM_BUG_ON(phys & ~PUD_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1394) set_pud(pudp, new_pud);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1395) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1396) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1397)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1398) int pmd_set_huge(pmd_t *pmdp, phys_addr_t phys, pgprot_t prot)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1399) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1400) pmd_t new_pmd = pfn_pmd(__phys_to_pfn(phys), mk_pmd_sect_prot(prot));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1401)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1402) /* Only allow permission changes for now */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1403) if (!pgattr_change_is_safe(READ_ONCE(pmd_val(*pmdp)),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1404) pmd_val(new_pmd)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1405) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1406)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1407) VM_BUG_ON(phys & ~PMD_MASK);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1408) set_pmd(pmdp, new_pmd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1409) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1410) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1411)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1412) int pud_clear_huge(pud_t *pudp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1413) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1414) if (!pud_sect(READ_ONCE(*pudp)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1415) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1416) pud_clear(pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1417) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1418) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1419)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1420) int pmd_clear_huge(pmd_t *pmdp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1421) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1422) if (!pmd_sect(READ_ONCE(*pmdp)))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1423) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1424) pmd_clear(pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1425) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1426) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1427)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1428) int pmd_free_pte_page(pmd_t *pmdp, unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1429) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1430) pte_t *table;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1431) pmd_t pmd;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1432)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1433) pmd = READ_ONCE(*pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1434)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1435) if (!pmd_table(pmd)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1436) VM_WARN_ON(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1437) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1438) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1439)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1440) table = pte_offset_kernel(pmdp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1441) pmd_clear(pmdp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1442) __flush_tlb_kernel_pgtable(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1443) pte_free_kernel(NULL, table);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1444) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1445) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1446)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1447) int pud_free_pmd_page(pud_t *pudp, unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1448) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1449) pmd_t *table;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1450) pmd_t *pmdp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1451) pud_t pud;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1452) unsigned long next, end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1453)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1454) pud = READ_ONCE(*pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1455)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1456) if (!pud_table(pud)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1457) VM_WARN_ON(1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1458) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1459) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1461) table = pmd_offset(pudp, addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1462) pmdp = table;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1463) next = addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1464) end = addr + PUD_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1465) do {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1466) pmd_free_pte_page(pmdp, next);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1467) } while (pmdp++, next += PMD_SIZE, next != end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1468)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1469) pud_clear(pudp);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1470) __flush_tlb_kernel_pgtable(addr);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1471) pmd_free(NULL, table);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1472) return 1;
^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) int p4d_free_pud_page(p4d_t *p4d, unsigned long addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1476) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1477) return 0; /* Don't attempt a block mapping */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1478) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1479)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1480) #ifdef CONFIG_MEMORY_HOTPLUG
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1481) static void __remove_pgd_mapping(pgd_t *pgdir, unsigned long start, u64 size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1482) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1483) unsigned long end = start + size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1484)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1485) WARN_ON(pgdir != init_mm.pgd);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1486) WARN_ON((start < PAGE_OFFSET) || (end > PAGE_END));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1487)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1488) unmap_hotplug_range(start, end, false, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1489) free_empty_tables(start, end, PAGE_OFFSET, PAGE_END);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1490) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1492) static bool inside_linear_region(u64 start, u64 size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1493) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1494) u64 start_linear_pa = __pa(_PAGE_OFFSET(vabits_actual));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1495) u64 end_linear_pa = __pa(PAGE_END - 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1496)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1497) if (IS_ENABLED(CONFIG_RANDOMIZE_BASE)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1498) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1499) * Check for a wrap, it is possible because of randomized linear
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1500) * mapping the start physical address is actually bigger than
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1501) * the end physical address. In this case set start to zero
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1502) * because [0, end_linear_pa] range must still be able to cover
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1503) * all addressable physical addresses.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1504) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1505) if (start_linear_pa > end_linear_pa)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1506) start_linear_pa = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1507) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1508)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1509) WARN_ON(start_linear_pa > end_linear_pa);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1510)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1511) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1512) * Linear mapping region is the range [PAGE_OFFSET..(PAGE_END - 1)]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1513) * accommodating both its ends but excluding PAGE_END. Max physical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1514) * range which can be mapped inside this linear mapping range, must
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1515) * also be derived from its end points.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1516) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1517) return start >= start_linear_pa && (start + size - 1) <= end_linear_pa;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1518) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1519)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1520) int arch_add_memory(int nid, u64 start, u64 size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1521) struct mhp_params *params)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1522) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1523) int ret, flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1524)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1525) if (!inside_linear_region(start, size)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1526) pr_err("[%llx %llx] is outside linear mapping region\n", start, start + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1527) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1528) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1529)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1530) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1531) * KFENCE requires linear map to be mapped at page granularity, so that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1532) * it is possible to protect/unprotect single pages in the KFENCE pool.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1533) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1534) if (rodata_full || debug_pagealloc_enabled() ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1535) IS_ENABLED(CONFIG_KFENCE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1536) flags = NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1537)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1538) __create_pgd_mapping(swapper_pg_dir, start, __phys_to_virt(start),
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1539) size, params->pgprot, __pgd_pgtable_alloc,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1540) flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1541)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1542) memblock_clear_nomap(start, size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1543)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1544) ret = __add_pages(nid, start >> PAGE_SHIFT, size >> PAGE_SHIFT,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1545) params);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1546) if (ret)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1547) __remove_pgd_mapping(swapper_pg_dir,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1548) __phys_to_virt(start), size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1549) else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1550) max_pfn = PFN_UP(start + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1551) max_low_pfn = max_pfn;
^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) return ret;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1555) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1556)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1557) void arch_remove_memory(int nid, u64 start, u64 size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1558) struct vmem_altmap *altmap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1559) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1560) unsigned long start_pfn = start >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1561) unsigned long nr_pages = size >> PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1562)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1563) __remove_pages(start_pfn, nr_pages, altmap);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1564) __remove_pgd_mapping(swapper_pg_dir, __phys_to_virt(start), size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1565) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1566)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1567) int check_range_driver_managed(u64 start, u64 size, const char *resource_name)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1568) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1569) struct mem_section *ms;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1570) unsigned long pfn = __phys_to_pfn(start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1571) unsigned long end_pfn = __phys_to_pfn(start + size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1572) struct resource *res;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1573) unsigned long flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1574)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1575) res = lookup_resource(&iomem_resource, start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1576) if (!res) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1577) pr_err("%s: couldn't find memory resource for start 0x%llx\n",
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1578) __func__, start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1579) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1580) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1581)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1582) flags = res->flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1583)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1584) if (!(flags & IORESOURCE_SYSRAM_DRIVER_MANAGED) ||
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1585) strstr(resource_name, "System RAM (") != resource_name)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1586) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1587)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1588) for (; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1589) ms = __pfn_to_section(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1590) if (early_section(ms))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1591) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1592) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1593)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1594) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1595) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1596)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1597) int populate_range_driver_managed(u64 start, u64 size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1598) const char *resource_name)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1599) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1600) unsigned long virt = (unsigned long)phys_to_virt(start);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1601) int flags = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1602)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1603) if (check_range_driver_managed(start, size, resource_name))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1604) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1605)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1606) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1607) * When rodata_full is enabled, memory is mapped at page size granule,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1608) * as opposed to block mapping.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1609) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1610) if (rodata_full || debug_pagealloc_enabled())
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1611) flags = NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1612)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1613) __create_pgd_mapping(init_mm.pgd, start, virt, size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1614) PAGE_KERNEL, NULL, flags);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1615)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1616) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1617) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1618) EXPORT_SYMBOL_GPL(populate_range_driver_managed);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1619)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1620) int depopulate_range_driver_managed(u64 start, u64 size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1621) const char *resource_name)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1622) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1623) if (check_range_driver_managed(start, size, resource_name))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1624) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1625)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1626) unmap_hotplug_range(start, start + size, false, NULL);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1627)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1628) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1629) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1630) EXPORT_SYMBOL_GPL(depopulate_range_driver_managed);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1631)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1632) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1633) * This memory hotplug notifier helps prevent boot memory from being
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1634) * inadvertently removed as it blocks pfn range offlining process in
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1635) * __offline_pages(). Hence this prevents both offlining as well as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1636) * removal process for boot memory which is initially always online.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1637) * In future if and when boot memory could be removed, this notifier
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1638) * should be dropped and free_hotplug_page_range() should handle any
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1639) * reserved pages allocated during boot.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1640) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1641) static int prevent_bootmem_remove_notifier(struct notifier_block *nb,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1642) unsigned long action, void *data)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1643) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1644) struct mem_section *ms;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1645) struct memory_notify *arg = data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1646) unsigned long end_pfn = arg->start_pfn + arg->nr_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1647) unsigned long pfn = arg->start_pfn;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1648)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1649) if (action != MEM_GOING_OFFLINE)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1650) return NOTIFY_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1651)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1652) for (; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1653) ms = __pfn_to_section(pfn);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1654) if (early_section(ms))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1655) return NOTIFY_BAD;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1656) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1657) return NOTIFY_OK;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1658) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1659)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1660) static struct notifier_block prevent_bootmem_remove_nb = {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1661) .notifier_call = prevent_bootmem_remove_notifier,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1662) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1663)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1664) static int __init prevent_bootmem_remove_init(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1665) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1666) return register_memory_notifier(&prevent_bootmem_remove_nb);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1667) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1668) device_initcall(prevent_bootmem_remove_init);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1669) #endif
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags