summaryrefslogtreecommitdiff
path: root/arch/x86/kernel/e820.c
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/kernel/e820.c')
-rw-r--r--arch/x86/kernel/e820.c120
1 files changed, 44 insertions, 76 deletions
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 4893d30ce438..57120f0749cc 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -28,18 +28,13 @@
* the first 128 E820 memory entries in boot_params.e820_table and the remaining
* (if any) entries of the SETUP_E820_EXT nodes. We use this to:
*
- * - inform the user about the firmware's notion of memory layout
- * via /sys/firmware/memmap
- *
* - the hibernation code uses it to generate a kernel-independent CRC32
* checksum of the physical memory layout of a system.
*
* - 'e820_table_kexec': a slightly modified (by the kernel) firmware version
* passed to us by the bootloader - the major difference between
- * e820_table_firmware[] and this one is that, the latter marks the setup_data
- * list created by the EFI boot stub as reserved, so that kexec can reuse the
- * setup_data information in the second kernel. Besides, e820_table_kexec[]
- * might also be modified by the kexec itself to fake a mptable.
+ * e820_table_firmware[] and this one is that e820_table_kexec[]
+ * might be modified by the kexec itself to fake an mptable.
* We use this to:
*
* - kexec, which is a bootloader in disguise, uses the original E820
@@ -47,6 +42,11 @@
* can have a restricted E820 map while the kexec()-ed kexec-kernel
* can have access to full memory - etc.
*
+ * Export the memory layout via /sys/firmware/memmap. kexec-tools uses
+ * the entries to create an E820 table for the kexec kernel.
+ *
+ * kexec_file_load in-kernel code uses the table for the kexec kernel.
+ *
* - 'e820_table': this is the main E820 table that is massaged by the
* low level x86 platform code, or modified by boot parameters, before
* passed on to higher level MM layers.
@@ -187,8 +187,7 @@ void __init e820__range_add(u64 start, u64 size, enum e820_type type)
static void __init e820_print_type(enum e820_type type)
{
switch (type) {
- case E820_TYPE_RAM: /* Fall through: */
- case E820_TYPE_RESERVED_KERN: pr_cont("usable"); break;
+ case E820_TYPE_RAM: pr_cont("usable"); break;
case E820_TYPE_RESERVED: pr_cont("reserved"); break;
case E820_TYPE_SOFT_RESERVED: pr_cont("soft reserved"); break;
case E820_TYPE_ACPI: pr_cont("ACPI data"); break;
@@ -764,7 +763,7 @@ void __init e820__register_nosave_regions(unsigned long limit_pfn)
pfn = PFN_DOWN(entry->addr + entry->size);
- if (entry->type != E820_TYPE_RAM && entry->type != E820_TYPE_RESERVED_KERN)
+ if (entry->type != E820_TYPE_RAM)
register_nosave_region(PFN_UP(entry->addr), pfn);
if (pfn >= limit_pfn)
@@ -991,60 +990,6 @@ static int __init parse_memmap_opt(char *str)
early_param("memmap", parse_memmap_opt);
/*
- * Reserve all entries from the bootloader's extensible data nodes list,
- * because if present we are going to use it later on to fetch e820
- * entries from it:
- */
-void __init e820__reserve_setup_data(void)
-{
- struct setup_indirect *indirect;
- struct setup_data *data;
- u64 pa_data, pa_next;
- u32 len;
-
- pa_data = boot_params.hdr.setup_data;
- if (!pa_data)
- return;
-
- while (pa_data) {
- data = early_memremap(pa_data, sizeof(*data));
- if (!data) {
- pr_warn("e820: failed to memremap setup_data entry\n");
- return;
- }
-
- len = sizeof(*data);
- pa_next = data->next;
-
- e820__range_update(pa_data, sizeof(*data)+data->len, E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
-
- if (data->type == SETUP_INDIRECT) {
- len += data->len;
- early_memunmap(data, sizeof(*data));
- data = early_memremap(pa_data, len);
- if (!data) {
- pr_warn("e820: failed to memremap indirect setup_data\n");
- return;
- }
-
- indirect = (struct setup_indirect *)data->data;
-
- if (indirect->type != SETUP_INDIRECT)
- e820__range_update(indirect->addr, indirect->len,
- E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
- }
-
- pa_data = pa_next;
- early_memunmap(data, len);
- }
-
- e820__update_table(e820_table);
-
- pr_info("extended physical RAM map:\n");
- e820__print_table("reserve setup_data");
-}
-
-/*
* Called after parse_early_param(), after early parameters (such as mem=)
* have been processed, in which case we already have an E820 table filled in
* via the parameter callback function(s), but it's not sorted and printed yet:
@@ -1063,7 +1008,6 @@ void __init e820__finish_early_params(void)
static const char *__init e820_type_to_string(struct e820_entry *entry)
{
switch (entry->type) {
- case E820_TYPE_RESERVED_KERN: /* Fall-through: */
case E820_TYPE_RAM: return "System RAM";
case E820_TYPE_ACPI: return "ACPI Tables";
case E820_TYPE_NVS: return "ACPI Non-volatile Storage";
@@ -1079,7 +1023,6 @@ static const char *__init e820_type_to_string(struct e820_entry *entry)
static unsigned long __init e820_type_to_iomem_type(struct e820_entry *entry)
{
switch (entry->type) {
- case E820_TYPE_RESERVED_KERN: /* Fall-through: */
case E820_TYPE_RAM: return IORESOURCE_SYSTEM_RAM;
case E820_TYPE_ACPI: /* Fall-through: */
case E820_TYPE_NVS: /* Fall-through: */
@@ -1101,7 +1044,6 @@ static unsigned long __init e820_type_to_iores_desc(struct e820_entry *entry)
case E820_TYPE_PRAM: return IORES_DESC_PERSISTENT_MEMORY_LEGACY;
case E820_TYPE_RESERVED: return IORES_DESC_RESERVED;
case E820_TYPE_SOFT_RESERVED: return IORES_DESC_SOFT_RESERVED;
- case E820_TYPE_RESERVED_KERN: /* Fall-through: */
case E820_TYPE_RAM: /* Fall-through: */
case E820_TYPE_UNUSABLE: /* Fall-through: */
default: return IORES_DESC_NONE;
@@ -1124,7 +1066,6 @@ static bool __init do_mark_busy(enum e820_type type, struct resource *res)
case E820_TYPE_PRAM:
case E820_TYPE_PMEM:
return false;
- case E820_TYPE_RESERVED_KERN:
case E820_TYPE_RAM:
case E820_TYPE_ACPI:
case E820_TYPE_NVS:
@@ -1146,11 +1087,8 @@ void __init e820__reserve_resources(void)
struct resource *res;
u64 end;
- res = memblock_alloc(sizeof(*res) * e820_table->nr_entries,
+ res = memblock_alloc_or_panic(sizeof(*res) * e820_table->nr_entries,
SMP_CACHE_BYTES);
- if (!res)
- panic("%s: Failed to allocate %zu bytes\n", __func__,
- sizeof(*res) * e820_table->nr_entries);
e820_res = res;
for (i = 0; i < e820_table->nr_entries; i++) {
@@ -1179,9 +1117,9 @@ void __init e820__reserve_resources(void)
res++;
}
- /* Expose the bootloader-provided memory layout to the sysfs. */
- for (i = 0; i < e820_table_firmware->nr_entries; i++) {
- struct e820_entry *entry = e820_table_firmware->entries + i;
+ /* Expose the kexec e820 table to the sysfs. */
+ for (i = 0; i < e820_table_kexec->nr_entries; i++) {
+ struct e820_entry *entry = e820_table_kexec->entries + i;
firmware_map_add_early(entry->addr, entry->addr + entry->size, e820_type_to_string(entry));
}
@@ -1305,6 +1243,36 @@ void __init e820__memblock_setup(void)
int i;
u64 end;
+#ifdef CONFIG_MEMORY_HOTPLUG
+ /*
+ * Memory used by the kernel cannot be hot-removed because Linux
+ * cannot migrate the kernel pages. When memory hotplug is
+ * enabled, we should prevent memblock from allocating memory
+ * for the kernel.
+ *
+ * ACPI SRAT records all hotpluggable memory ranges. But before
+ * SRAT is parsed, we don't know about it.
+ *
+ * The kernel image is loaded into memory at very early time. We
+ * cannot prevent this anyway. So on NUMA system, we set any
+ * node the kernel resides in as un-hotpluggable.
+ *
+ * Since on modern servers, one node could have double-digit
+ * gigabytes memory, we can assume the memory around the kernel
+ * image is also un-hotpluggable. So before SRAT is parsed, just
+ * allocate memory near the kernel image to try the best to keep
+ * the kernel away from hotpluggable memory.
+ */
+ if (movable_node_is_enabled())
+ memblock_set_bottom_up(true);
+#endif
+
+ /*
+ * At this point only the first megabyte is mapped for sure, the
+ * rest of the memory cannot be used for memblock resizing
+ */
+ memblock_set_current_limit(ISA_END_ADDRESS);
+
/*
* The bootstrap memblock region count maximum is 128 entries
* (INIT_MEMBLOCK_REGIONS), but EFI might pass us more E820 entries
@@ -1326,7 +1294,7 @@ void __init e820__memblock_setup(void)
if (entry->type == E820_TYPE_SOFT_RESERVED)
memblock_reserve(entry->addr, entry->size);
- if (entry->type != E820_TYPE_RAM && entry->type != E820_TYPE_RESERVED_KERN)
+ if (entry->type != E820_TYPE_RAM)
continue;
memblock_add(entry->addr, entry->size);