^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) // SPDX-License-Identifier: GPL-2.0
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) * kaslr.c
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5) * This contains the routines needed to generate a reasonable level of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) * entropy to choose a randomized kernel base address offset in support
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) * of Kernel Address Space Layout Randomization (KASLR). Additionally
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8) * handles walking the physical memory maps (and tracking memory regions
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) * to avoid) in order to select a physical memory location that can
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) * contain the entire properly aligned running kernel image.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) * isspace() in linux/ctype.h is expected by next_args() to filter
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16) * out "space/lf/tab". While boot/ctype.h conflicts with linux/ctype.h,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) * since isdigit() is implemented in both of them. Hence disable it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) * here.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) #define BOOT_CTYPE_H
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) #include "misc.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) #include "error.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) #include "../string.h"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) #include <generated/compile.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27) #include <linux/module.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28) #include <linux/uts.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29) #include <linux/utsname.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) #include <linux/ctype.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) #include <linux/efi.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32) #include <generated/utsrelease.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) #include <asm/efi.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) /* Macros used by the included decompressor code below. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36) #define STATIC
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) #include <linux/decompress/mm.h>
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) #define _SETUP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40) #include <asm/setup.h> /* For COMMAND_LINE_SIZE */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) #undef _SETUP
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) extern unsigned long get_cmd_line_ptr(void);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) /* Simplified build-specific string for starting entropy. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46) static const char build_str[] = UTS_RELEASE " (" LINUX_COMPILE_BY "@"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) LINUX_COMPILE_HOST ") (" LINUX_COMPILER ") " UTS_VERSION;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) static unsigned long rotate_xor(unsigned long hash, const void *area,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50) size_t size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52) size_t i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) unsigned long *ptr = (unsigned long *)area;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55) for (i = 0; i < size / sizeof(hash); i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) /* Rotate by odd number of bits and XOR. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) hash = (hash << ((sizeof(hash) * 8) - 7)) | (hash >> 7);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) hash ^= ptr[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) return hash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) /* Attempt to create a simple but unpredictable starting entropy. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) static unsigned long get_boot_seed(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) unsigned long hash = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) hash = rotate_xor(hash, build_str, sizeof(build_str));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70) hash = rotate_xor(hash, boot_params, sizeof(*boot_params));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) return hash;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) #define KASLR_COMPRESSED_BOOT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) #include "../../lib/kaslr.c"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) /* Only supporting at most 4 unusable memmap regions with kaslr */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) #define MAX_MEMMAP_REGIONS 4
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) static bool memmap_too_large;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86) * Store memory limit: MAXMEM on 64-bit and KERNEL_IMAGE_SIZE on 32-bit.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87) * It may be reduced by "mem=nn[KMG]" or "memmap=nn[KMG]" command line options.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) static u64 mem_limit;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91) /* Number of immovable memory regions */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) static int num_immovable_mem;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) enum mem_avoid_index {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) MEM_AVOID_ZO_RANGE = 0,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96) MEM_AVOID_INITRD,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) MEM_AVOID_CMDLINE,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98) MEM_AVOID_BOOTPARAMS,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) MEM_AVOID_MEMMAP_BEGIN,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) MEM_AVOID_MEMMAP_END = MEM_AVOID_MEMMAP_BEGIN + MAX_MEMMAP_REGIONS - 1,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) MEM_AVOID_MAX,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) static struct mem_vector mem_avoid[MEM_AVOID_MAX];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) static bool mem_overlaps(struct mem_vector *one, struct mem_vector *two)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) /* Item one is entirely before item two. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) if (one->start + one->size <= two->start)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111) /* Item one is entirely after item two. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) if (one->start >= two->start + two->size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117) char *skip_spaces(const char *str)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) while (isspace(*str))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) ++str;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121) return (char *)str;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123) #include "../../../../lib/ctype.c"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) #include "../../../../lib/cmdline.c"
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126) enum parse_mode {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) PARSE_MEMMAP,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) PARSE_EFI,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131) static int
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) parse_memmap(char *p, u64 *start, u64 *size, enum parse_mode mode)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) char *oldp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) if (!p)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) /* We don't care about this option here */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) if (!strncmp(p, "exactmap", 8))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143) oldp = p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) *size = memparse(p, &p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) if (p == oldp)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) switch (*p) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) case '#':
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) case '$':
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151) case '!':
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) *start = memparse(p + 1, &p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) case '@':
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) if (mode == PARSE_MEMMAP) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) * memmap=nn@ss specifies usable region, should
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158) * be skipped
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) *size = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162) u64 flags;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) * efi_fake_mem=nn@ss:attr the attr specifies
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166) * flags that might imply a soft-reservation.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) *start = memparse(p + 1, &p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) if (p && *p == ':') {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170) p++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) if (kstrtoull(p, 0, &flags) < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172) *size = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) else if (flags & EFI_MEMORY_SP)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176) *size = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178) fallthrough;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) default:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) * If w/o offset, only size specified, memmap=nn[KMG] has the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182) * same behaviour as mem=nn[KMG]. It limits the max address
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) * system can use. Region above the limit should be avoided.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) *start = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189) return -EINVAL;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) static void mem_avoid_memmap(enum parse_mode mode, char *str)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) static int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196) if (i >= MAX_MEMMAP_REGIONS)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) while (str && (i < MAX_MEMMAP_REGIONS)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) int rc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201) u64 start, size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202) char *k = strchr(str, ',');
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204) if (k)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) *k++ = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207) rc = parse_memmap(str, &start, &size, mode);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) if (rc < 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) str = k;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) if (start == 0) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213) /* Store the specified memory limit if size > 0 */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) if (size > 0 && size < mem_limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) mem_limit = size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) mem_avoid[MEM_AVOID_MEMMAP_BEGIN + i].start = start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) mem_avoid[MEM_AVOID_MEMMAP_BEGIN + i].size = size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222) i++;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) /* More than 4 memmaps, fail kaslr */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226) if ((i >= MAX_MEMMAP_REGIONS) && str)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) memmap_too_large = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230) /* Store the number of 1GB huge pages which users specified: */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) static unsigned long max_gb_huge_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) static void parse_gb_huge_pages(char *param, char *val)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235) static bool gbpage_sz;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) char *p;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238) if (!strcmp(param, "hugepagesz")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) p = val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) if (memparse(p, &p) != PUD_SIZE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241) gbpage_sz = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) if (gbpage_sz)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) warn("Repeatedly set hugeTLB page size of 1G!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247) gbpage_sz = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 250)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 251) if (!strcmp(param, "hugepages") && gbpage_sz) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 252) p = val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 253) max_gb_huge_pages = simple_strtoull(p, &p, 0);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 254) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 255) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 256) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 257)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 258) static void handle_mem_options(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 259) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 260) char *args = (char *)get_cmd_line_ptr();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 261) size_t len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 262) char *tmp_cmdline;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 263) char *param, *val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 264) u64 mem_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 265)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 266) if (!args)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 267) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 268)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 269) len = strnlen(args, COMMAND_LINE_SIZE-1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 270) tmp_cmdline = malloc(len + 1);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 271) if (!tmp_cmdline)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 272) error("Failed to allocate space for tmp_cmdline");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 273)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 274) memcpy(tmp_cmdline, args, len);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 275) tmp_cmdline[len] = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 276) args = tmp_cmdline;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 277)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 278) /* Chew leading spaces */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 279) args = skip_spaces(args);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 280)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 281) while (*args) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 282) args = next_arg(args, ¶m, &val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 283) /* Stop at -- */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 284) if (!val && strcmp(param, "--") == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 285) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 286)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 287) if (!strcmp(param, "memmap")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 288) mem_avoid_memmap(PARSE_MEMMAP, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 289) } else if (IS_ENABLED(CONFIG_X86_64) && strstr(param, "hugepages")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 290) parse_gb_huge_pages(param, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 291) } else if (!strcmp(param, "mem")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 292) char *p = val;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 293)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 294) if (!strcmp(p, "nopentium"))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 295) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 296) mem_size = memparse(p, &p);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 297) if (mem_size == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 298) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 299)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 300) if (mem_size < mem_limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 301) mem_limit = mem_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 302) } else if (!strcmp(param, "efi_fake_mem")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 303) mem_avoid_memmap(PARSE_EFI, val);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 304) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 305) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 306)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 307) free(tmp_cmdline);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 308) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 309) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 310)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 311) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 312) * In theory, KASLR can put the kernel anywhere in the range of [16M, MAXMEM)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 313) * on 64-bit, and [16M, KERNEL_IMAGE_SIZE) on 32-bit.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 314) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 315) * The mem_avoid array is used to store the ranges that need to be avoided
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 316) * when KASLR searches for an appropriate random address. We must avoid any
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 317) * regions that are unsafe to overlap with during decompression, and other
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 318) * things like the initrd, cmdline and boot_params. This comment seeks to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 319) * explain mem_avoid as clearly as possible since incorrect mem_avoid
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 320) * memory ranges lead to really hard to debug boot failures.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 321) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 322) * The initrd, cmdline, and boot_params are trivial to identify for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 323) * avoiding. They are MEM_AVOID_INITRD, MEM_AVOID_CMDLINE, and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 324) * MEM_AVOID_BOOTPARAMS respectively below.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 325) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 326) * What is not obvious how to avoid is the range of memory that is used
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 327) * during decompression (MEM_AVOID_ZO_RANGE below). This range must cover
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 328) * the compressed kernel (ZO) and its run space, which is used to extract
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 329) * the uncompressed kernel (VO) and relocs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 330) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 331) * ZO's full run size sits against the end of the decompression buffer, so
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 332) * we can calculate where text, data, bss, etc of ZO are positioned more
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 333) * easily.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 334) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 335) * For additional background, the decompression calculations can be found
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 336) * in header.S, and the memory diagram is based on the one found in misc.c.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 337) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 338) * The following conditions are already enforced by the image layouts and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 339) * associated code:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 340) * - input + input_size >= output + output_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 341) * - kernel_total_size <= init_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 342) * - kernel_total_size <= output_size (see Note below)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 343) * - output + init_size >= output + output_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 344) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 345) * (Note that kernel_total_size and output_size have no fundamental
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 346) * relationship, but output_size is passed to choose_random_location
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 347) * as a maximum of the two. The diagram is showing a case where
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 348) * kernel_total_size is larger than output_size, but this case is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 349) * handled by bumping output_size.)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 350) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 351) * The above conditions can be illustrated by a diagram:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 352) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 353) * 0 output input input+input_size output+init_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 354) * | | | | |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 355) * | | | | |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 356) * |-----|--------|--------|--------------|-----------|--|-------------|
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 357) * | | |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 358) * | | |
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 359) * output+init_size-ZO_INIT_SIZE output+output_size output+kernel_total_size
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 360) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 361) * [output, output+init_size) is the entire memory range used for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 362) * extracting the compressed image.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 363) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 364) * [output, output+kernel_total_size) is the range needed for the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 365) * uncompressed kernel (VO) and its run size (bss, brk, etc).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 366) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 367) * [output, output+output_size) is VO plus relocs (i.e. the entire
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 368) * uncompressed payload contained by ZO). This is the area of the buffer
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 369) * written to during decompression.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 370) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 371) * [output+init_size-ZO_INIT_SIZE, output+init_size) is the worst-case
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 372) * range of the copied ZO and decompression code. (i.e. the range
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 373) * covered backwards of size ZO_INIT_SIZE, starting from output+init_size.)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 374) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 375) * [input, input+input_size) is the original copied compressed image (ZO)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 376) * (i.e. it does not include its run size). This range must be avoided
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 377) * because it contains the data used for decompression.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 378) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 379) * [input+input_size, output+init_size) is [_text, _end) for ZO. This
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 380) * range includes ZO's heap and stack, and must be avoided since it
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 381) * performs the decompression.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 382) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 383) * Since the above two ranges need to be avoided and they are adjacent,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 384) * they can be merged, resulting in: [input, output+init_size) which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 385) * becomes the MEM_AVOID_ZO_RANGE below.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 386) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 387) static void mem_avoid_init(unsigned long input, unsigned long input_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 388) unsigned long output)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 389) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 390) unsigned long init_size = boot_params->hdr.init_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 391) u64 initrd_start, initrd_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 392) unsigned long cmd_line, cmd_line_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 393)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 394) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 395) * Avoid the region that is unsafe to overlap during
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 396) * decompression.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 397) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 398) mem_avoid[MEM_AVOID_ZO_RANGE].start = input;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 399) mem_avoid[MEM_AVOID_ZO_RANGE].size = (output + init_size) - input;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 400)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 401) /* Avoid initrd. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 402) initrd_start = (u64)boot_params->ext_ramdisk_image << 32;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 403) initrd_start |= boot_params->hdr.ramdisk_image;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 404) initrd_size = (u64)boot_params->ext_ramdisk_size << 32;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 405) initrd_size |= boot_params->hdr.ramdisk_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 406) mem_avoid[MEM_AVOID_INITRD].start = initrd_start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 407) mem_avoid[MEM_AVOID_INITRD].size = initrd_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 408) /* No need to set mapping for initrd, it will be handled in VO. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 409)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 410) /* Avoid kernel command line. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 411) cmd_line = get_cmd_line_ptr();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 412) /* Calculate size of cmd_line. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 413) if (cmd_line) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 414) cmd_line_size = strnlen((char *)cmd_line, COMMAND_LINE_SIZE-1) + 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 415) mem_avoid[MEM_AVOID_CMDLINE].start = cmd_line;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 416) mem_avoid[MEM_AVOID_CMDLINE].size = cmd_line_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 417) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 418)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 419) /* Avoid boot parameters. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 420) mem_avoid[MEM_AVOID_BOOTPARAMS].start = (unsigned long)boot_params;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 421) mem_avoid[MEM_AVOID_BOOTPARAMS].size = sizeof(*boot_params);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 422)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 423) /* We don't need to set a mapping for setup_data. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 424)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 425) /* Mark the memmap regions we need to avoid */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 426) handle_mem_options();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 427)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 428) /* Enumerate the immovable memory regions */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 429) num_immovable_mem = count_immovable_mem_regions();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 430) }
^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) * Does this memory vector overlap a known avoided area? If so, record the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 434) * overlap region with the lowest address.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 435) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 436) static bool mem_avoid_overlap(struct mem_vector *img,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 437) struct mem_vector *overlap)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 438) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 439) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 440) struct setup_data *ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 441) u64 earliest = img->start + img->size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 442) bool is_overlapping = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 443)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 444) for (i = 0; i < MEM_AVOID_MAX; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 445) if (mem_overlaps(img, &mem_avoid[i]) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 446) mem_avoid[i].start < earliest) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 447) *overlap = mem_avoid[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 448) earliest = overlap->start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 449) is_overlapping = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 450) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 451) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 452)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 453) /* Avoid all entries in the setup_data linked list. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 454) ptr = (struct setup_data *)(unsigned long)boot_params->hdr.setup_data;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 455) while (ptr) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 456) struct mem_vector avoid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 457)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 458) avoid.start = (unsigned long)ptr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 459) avoid.size = sizeof(*ptr) + ptr->len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 460)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 461) if (mem_overlaps(img, &avoid) && (avoid.start < earliest)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 462) *overlap = avoid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 463) earliest = overlap->start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 464) is_overlapping = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 465) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 466)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 467) if (ptr->type == SETUP_INDIRECT &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 468) ((struct setup_indirect *)ptr->data)->type != SETUP_INDIRECT) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 469) avoid.start = ((struct setup_indirect *)ptr->data)->addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 470) avoid.size = ((struct setup_indirect *)ptr->data)->len;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 471)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 472) if (mem_overlaps(img, &avoid) && (avoid.start < earliest)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 473) *overlap = avoid;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 474) earliest = overlap->start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 475) is_overlapping = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 476) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 477) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 478)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 479) ptr = (struct setup_data *)(unsigned long)ptr->next;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 480) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 481)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 482) return is_overlapping;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 483) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 484)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 485) struct slot_area {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 486) u64 addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 487) unsigned long num;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 488) };
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 489)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 490) #define MAX_SLOT_AREA 100
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 491)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 492) static struct slot_area slot_areas[MAX_SLOT_AREA];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 493) static unsigned int slot_area_index;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 494) static unsigned long slot_max;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 495)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 496) static void store_slot_info(struct mem_vector *region, unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 497) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 498) struct slot_area slot_area;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 499)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 500) if (slot_area_index == MAX_SLOT_AREA)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 501) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 502)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 503) slot_area.addr = region->start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 504) slot_area.num = 1 + (region->size - image_size) / CONFIG_PHYSICAL_ALIGN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 505)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 506) slot_areas[slot_area_index++] = slot_area;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 507) slot_max += slot_area.num;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 508) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 509)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 510) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 511) * Skip as many 1GB huge pages as possible in the passed region
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 512) * according to the number which users specified:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 513) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 514) static void
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 515) process_gb_huge_pages(struct mem_vector *region, unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 516) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 517) u64 pud_start, pud_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 518) unsigned long gb_huge_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 519) struct mem_vector tmp;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 520)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 521) if (!IS_ENABLED(CONFIG_X86_64) || !max_gb_huge_pages) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 522) store_slot_info(region, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 523) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 524) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 525)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 526) /* Are there any 1GB pages in the region? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 527) pud_start = ALIGN(region->start, PUD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 528) pud_end = ALIGN_DOWN(region->start + region->size, PUD_SIZE);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 529)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 530) /* No good 1GB huge pages found: */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 531) if (pud_start >= pud_end) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 532) store_slot_info(region, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 533) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 534) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 535)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 536) /* Check if the head part of the region is usable. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 537) if (pud_start >= region->start + image_size) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 538) tmp.start = region->start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 539) tmp.size = pud_start - region->start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 540) store_slot_info(&tmp, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 541) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 542)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 543) /* Skip the good 1GB pages. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 544) gb_huge_pages = (pud_end - pud_start) >> PUD_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 545) if (gb_huge_pages > max_gb_huge_pages) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 546) pud_end = pud_start + (max_gb_huge_pages << PUD_SHIFT);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 547) max_gb_huge_pages = 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 548) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 549) max_gb_huge_pages -= gb_huge_pages;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 550) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 551)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 552) /* Check if the tail part of the region is usable. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 553) if (region->start + region->size >= pud_end + image_size) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 554) tmp.start = pud_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 555) tmp.size = region->start + region->size - pud_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 556) store_slot_info(&tmp, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 557) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 558) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 559)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 560) static u64 slots_fetch_random(void)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 561) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 562) unsigned long slot;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 563) unsigned int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 564)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 565) /* Handle case of no slots stored. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 566) if (slot_max == 0)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 567) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 568)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 569) slot = kaslr_get_random_long("Physical") % slot_max;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 570)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 571) for (i = 0; i < slot_area_index; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 572) if (slot >= slot_areas[i].num) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 573) slot -= slot_areas[i].num;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 574) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 575) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 576) return slot_areas[i].addr + ((u64)slot * CONFIG_PHYSICAL_ALIGN);
^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) if (i == slot_area_index)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 580) debug_putstr("slots_fetch_random() failed!?\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 581) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 582) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 583)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 584) static void __process_mem_region(struct mem_vector *entry,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 585) unsigned long minimum,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 586) unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 587) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 588) struct mem_vector region, overlap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 589) u64 region_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 590)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 591) /* Enforce minimum and memory limit. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 592) region.start = max_t(u64, entry->start, minimum);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 593) region_end = min(entry->start + entry->size, mem_limit);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 594)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 595) /* Give up if slot area array is full. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 596) while (slot_area_index < MAX_SLOT_AREA) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 597) /* Potentially raise address to meet alignment needs. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 598) region.start = ALIGN(region.start, CONFIG_PHYSICAL_ALIGN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 599)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 600) /* Did we raise the address above the passed in memory entry? */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 601) if (region.start > region_end)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 602) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 603)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 604) /* Reduce size by any delta from the original address. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 605) region.size = region_end - region.start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 606)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 607) /* Return if region can't contain decompressed kernel */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 608) if (region.size < image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 609) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 610)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 611) /* If nothing overlaps, store the region and return. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 612) if (!mem_avoid_overlap(®ion, &overlap)) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 613) process_gb_huge_pages(®ion, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 614) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 615) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 616)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 617) /* Store beginning of region if holds at least image_size. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 618) if (overlap.start >= region.start + image_size) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 619) region.size = overlap.start - region.start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 620) process_gb_huge_pages(®ion, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 621) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 622)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 623) /* Clip off the overlapping region and start over. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 624) region.start = overlap.start + overlap.size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 625) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 626) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 627)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 628) static bool process_mem_region(struct mem_vector *region,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 629) unsigned long minimum,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 630) unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 631) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 632) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 633) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 634) * If no immovable memory found, or MEMORY_HOTREMOVE disabled,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 635) * use @region directly.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 636) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 637) if (!num_immovable_mem) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 638) __process_mem_region(region, minimum, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 639)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 640) if (slot_area_index == MAX_SLOT_AREA) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 641) debug_putstr("Aborted e820/efi memmap scan (slot_areas full)!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 642) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 643) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 644) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 645) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 646)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 647) #if defined(CONFIG_MEMORY_HOTREMOVE) && defined(CONFIG_ACPI)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 648) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 649) * If immovable memory found, filter the intersection between
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 650) * immovable memory and @region.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 651) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 652) for (i = 0; i < num_immovable_mem; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 653) u64 start, end, entry_end, region_end;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 654) struct mem_vector entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 655)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 656) if (!mem_overlaps(region, &immovable_mem[i]))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 657) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 658)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 659) start = immovable_mem[i].start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 660) end = start + immovable_mem[i].size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 661) region_end = region->start + region->size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 662)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 663) entry.start = clamp(region->start, start, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 664) entry_end = clamp(region_end, start, end);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 665) entry.size = entry_end - entry.start;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 666)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 667) __process_mem_region(&entry, minimum, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 668)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 669) if (slot_area_index == MAX_SLOT_AREA) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 670) debug_putstr("Aborted e820/efi memmap scan when walking immovable regions(slot_areas full)!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 671) return 1;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 672) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 673) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 674) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 675) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 676) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 677)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 678) #ifdef CONFIG_EFI
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 679) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 680) * Returns true if we processed the EFI memmap, which we prefer over the E820
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 681) * table if it is available.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 682) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 683) static bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 684) process_efi_entries(unsigned long minimum, unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 685) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 686) struct efi_info *e = &boot_params->efi_info;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 687) bool efi_mirror_found = false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 688) struct mem_vector region;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 689) efi_memory_desc_t *md;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 690) unsigned long pmap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 691) char *signature;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 692) u32 nr_desc;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 693) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 694)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 695) signature = (char *)&e->efi_loader_signature;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 696) if (strncmp(signature, EFI32_LOADER_SIGNATURE, 4) &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 697) strncmp(signature, EFI64_LOADER_SIGNATURE, 4))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 698) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 699)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 700) #ifdef CONFIG_X86_32
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 701) /* Can't handle data above 4GB at this time */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 702) if (e->efi_memmap_hi) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 703) warn("EFI memmap is above 4GB, can't be handled now on x86_32. EFI should be disabled.\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 704) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 705) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 706) pmap = e->efi_memmap;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 707) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 708) pmap = (e->efi_memmap | ((__u64)e->efi_memmap_hi << 32));
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 709) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 710)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 711) nr_desc = e->efi_memmap_size / e->efi_memdesc_size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 712) for (i = 0; i < nr_desc; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 713) md = efi_early_memdesc_ptr(pmap, e->efi_memdesc_size, i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 714) if (md->attribute & EFI_MEMORY_MORE_RELIABLE) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 715) efi_mirror_found = true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 716) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 717) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 718) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 719)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 720) for (i = 0; i < nr_desc; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 721) md = efi_early_memdesc_ptr(pmap, e->efi_memdesc_size, i);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 722)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 723) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 724) * Here we are more conservative in picking free memory than
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 725) * the EFI spec allows:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 726) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 727) * According to the spec, EFI_BOOT_SERVICES_{CODE|DATA} are also
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 728) * free memory and thus available to place the kernel image into,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 729) * but in practice there's firmware where using that memory leads
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 730) * to crashes.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 731) *
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 732) * Only EFI_CONVENTIONAL_MEMORY is guaranteed to be free.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 733) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 734) if (md->type != EFI_CONVENTIONAL_MEMORY)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 735) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 736)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 737) if (efi_soft_reserve_enabled() &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 738) (md->attribute & EFI_MEMORY_SP))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 739) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 740)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 741) if (efi_mirror_found &&
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 742) !(md->attribute & EFI_MEMORY_MORE_RELIABLE))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 743) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 744)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 745) region.start = md->phys_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 746) region.size = md->num_pages << EFI_PAGE_SHIFT;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 747) if (process_mem_region(®ion, minimum, image_size))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 748) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 749) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 750) return true;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 751) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 752) #else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 753) static inline bool
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 754) process_efi_entries(unsigned long minimum, unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 755) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 756) return false;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 757) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 758) #endif
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 759)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 760) static void process_e820_entries(unsigned long minimum,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 761) unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 762) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 763) int i;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 764) struct mem_vector region;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 765) struct boot_e820_entry *entry;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 766)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 767) /* Verify potential e820 positions, appending to slots list. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 768) for (i = 0; i < boot_params->e820_entries; i++) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 769) entry = &boot_params->e820_table[i];
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 770) /* Skip non-RAM entries. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 771) if (entry->type != E820_TYPE_RAM)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 772) continue;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 773) region.start = entry->addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 774) region.size = entry->size;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 775) if (process_mem_region(®ion, minimum, image_size))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 776) break;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 777) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 778) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 779)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 780) static unsigned long find_random_phys_addr(unsigned long minimum,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 781) unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 782) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 783) u64 phys_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 784)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 785) /* Bail out early if it's impossible to succeed. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 786) if (minimum + image_size > mem_limit)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 787) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 788)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 789) /* Check if we had too many memmaps. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 790) if (memmap_too_large) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 791) debug_putstr("Aborted memory entries scan (more than 4 memmap= args)!\n");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 792) return 0;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 793) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 794)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 795) if (!process_efi_entries(minimum, image_size))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 796) process_e820_entries(minimum, image_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 797)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 798) phys_addr = slots_fetch_random();
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 799)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 800) /* Perform a final check to make sure the address is in range. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 801) if (phys_addr < minimum || phys_addr + image_size > mem_limit) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 802) warn("Invalid physical address chosen!\n");
^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)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 806) return (unsigned long)phys_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 807) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 808)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 809) static unsigned long find_random_virt_addr(unsigned long minimum,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 810) unsigned long image_size)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 811) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 812) unsigned long slots, random_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 813)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 814) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 815) * There are how many CONFIG_PHYSICAL_ALIGN-sized slots
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 816) * that can hold image_size within the range of minimum to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 817) * KERNEL_IMAGE_SIZE?
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 818) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 819) slots = 1 + (KERNEL_IMAGE_SIZE - minimum - image_size) / CONFIG_PHYSICAL_ALIGN;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 820)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 821) random_addr = kaslr_get_random_long("Virtual") % slots;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 822)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 823) return random_addr * CONFIG_PHYSICAL_ALIGN + minimum;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 824) }
^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) * Since this function examines addresses much more numerically,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 828) * it takes the input and output pointers as 'unsigned long'.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 829) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 830) void choose_random_location(unsigned long input,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 831) unsigned long input_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 832) unsigned long *output,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 833) unsigned long output_size,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 834) unsigned long *virt_addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 835) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 836) unsigned long random_addr, min_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 837)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 838) if (cmdline_find_option_bool("nokaslr")) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 839) warn("KASLR disabled: 'nokaslr' on cmdline.");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 840) return;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 841) }
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 842)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 843) boot_params->hdr.loadflags |= KASLR_FLAG;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 844)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 845) if (IS_ENABLED(CONFIG_X86_32))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 846) mem_limit = KERNEL_IMAGE_SIZE;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 847) else
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 848) mem_limit = MAXMEM;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 849)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 850) /* Record the various known unsafe memory ranges. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 851) mem_avoid_init(input, input_size, *output);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 852)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 853) /*
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 854) * Low end of the randomization range should be the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 855) * smaller of 512M or the initial kernel image
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 856) * location:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 857) */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 858) min_addr = min(*output, 512UL << 20);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 859) /* Make sure minimum is aligned. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 860) min_addr = ALIGN(min_addr, CONFIG_PHYSICAL_ALIGN);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 861)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 862) /* Walk available memory entries to find a random address. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 863) random_addr = find_random_phys_addr(min_addr, output_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 864) if (!random_addr) {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 865) warn("Physical KASLR disabled: no suitable memory region!");
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 866) } else {
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 867) /* Update the new physical address location. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 868) if (*output != random_addr)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 869) *output = random_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 870) }
^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) /* Pick random virtual address starting from LOAD_PHYSICAL_ADDR. */
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 874) if (IS_ENABLED(CONFIG_X86_64))
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 875) random_addr = find_random_virt_addr(LOAD_PHYSICAL_ADDR, output_size);
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 876) *virt_addr = random_addr;
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 877) }
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags