summaryrefslogtreecommitdiff
path: root/mm/memblock.c
diff options
context:
space:
mode:
Diffstat (limited to 'mm/memblock.c')
-rw-r--r--mm/memblock.c463
1 files changed, 341 insertions, 122 deletions
diff --git a/mm/memblock.c b/mm/memblock.c
index b3ddfdec7a80..021db49eb7fc 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -17,17 +17,17 @@
#include <linux/seq_file.h>
#include <linux/memblock.h>
#include <linux/mutex.h>
+#include <linux/string_helpers.h>
-#ifdef CONFIG_KEXEC_HANDOVER
#include <linux/libfdt.h>
#include <linux/kexec_handover.h>
#include <linux/kho/abi/memblock.h>
-#endif /* CONFIG_KEXEC_HANDOVER */
#include <asm/sections.h>
#include <linux/io.h>
#include "internal.h"
+#include "mm_init.h"
#define INIT_MEMBLOCK_REGIONS 128
#define INIT_PHYSMEM_REGIONS 4
@@ -384,26 +384,27 @@ static void __init_memblock memblock_remove_region(struct memblock_type *type, u
*/
void __init memblock_discard(void)
{
- phys_addr_t addr, size;
+ phys_addr_t size;
+ void *addr;
if (memblock.reserved.regions != memblock_reserved_init_regions) {
- addr = __pa(memblock.reserved.regions);
+ addr = memblock.reserved.regions;
size = PAGE_ALIGN(sizeof(struct memblock_region) *
memblock.reserved.max);
if (memblock_reserved_in_slab)
- kfree(memblock.reserved.regions);
+ kfree(addr);
else
- memblock_free_late(addr, size);
+ memblock_free(addr, size);
}
if (memblock.memory.regions != memblock_memory_init_regions) {
- addr = __pa(memblock.memory.regions);
+ addr = memblock.memory.regions;
size = PAGE_ALIGN(sizeof(struct memblock_region) *
memblock.memory.max);
if (memblock_memory_in_slab)
- kfree(memblock.memory.regions);
+ kfree(addr);
else
- memblock_free_late(addr, size);
+ memblock_free(addr, size);
}
memblock_memory = NULL;
@@ -893,13 +894,81 @@ int __init_memblock memblock_remove(phys_addr_t base, phys_addr_t size)
return memblock_remove_range(&memblock.memory, base, size);
}
+static unsigned long __free_reserved_area(phys_addr_t start, phys_addr_t end,
+ int poison)
+{
+ unsigned long pages = 0, pfn;
+
+ if (deferred_pages_enabled()) {
+ WARN(1, "Cannot free reserved memory because of deferred initialization of the memory map");
+ return 0;
+ }
+
+ for_each_valid_pfn(pfn, PFN_UP(start), PFN_DOWN(end)) {
+ struct page *page = pfn_to_page(pfn);
+ void *direct_map_addr;
+
+ /*
+ * 'direct_map_addr' might be different from the kernel virtual
+ * address because some architectures use aliases.
+ * Going via physical address, pfn_to_page() and page_address()
+ * ensures that we get a _writeable_ alias for the memset().
+ */
+ direct_map_addr = page_address(page);
+ /*
+ * Perform a kasan-unchecked memset() since this memory
+ * has not been initialized.
+ */
+ direct_map_addr = kasan_reset_tag(direct_map_addr);
+ if ((unsigned int)poison <= 0xFF)
+ memset(direct_map_addr, poison, PAGE_SIZE);
+
+ free_reserved_page(page);
+ pages++;
+ }
+ return pages;
+}
+
+unsigned long free_reserved_area(void *start, void *end, int poison, const char *s)
+{
+ phys_addr_t start_pa, end_pa;
+ unsigned long pages;
+
+ /*
+ * end is the first address past the region and it may be beyond what
+ * __pa() or __pa_symbol() can handle.
+ * Use the address included in the range for the conversion and add back
+ * 1 afterwards.
+ */
+ if (__is_kernel((unsigned long)start)) {
+ start_pa = __pa_symbol(start);
+ end_pa = __pa_symbol(end - 1) + 1;
+ } else {
+ start_pa = __pa(start);
+ end_pa = __pa(end - 1) + 1;
+ }
+
+ if (IS_ENABLED(CONFIG_ARCH_KEEP_MEMBLOCK)) {
+ if (start_pa < end_pa)
+ memblock_remove_range(&memblock.reserved,
+ start_pa, end_pa - start_pa);
+ }
+
+ pages = __free_reserved_area(start_pa, end_pa, poison);
+ if (pages && s)
+ pr_info("Freeing %s memory: %ldK\n", s, K(pages));
+
+ return pages;
+}
+
/**
* memblock_free - free boot memory allocation
* @ptr: starting address of the boot memory allocation
* @size: size of the boot memory block in bytes
*
* Free boot memory block previously allocated by memblock_alloc_xx() API.
- * The freeing memory will not be released to the buddy allocator.
+ * If called after the buddy allocator is available, the memory is released to
+ * the buddy allocator.
*/
void __init_memblock memblock_free(void *ptr, size_t size)
{
@@ -913,17 +982,26 @@ void __init_memblock memblock_free(void *ptr, size_t size)
* @size: size of the boot memory block in bytes
*
* Free boot memory block previously allocated by memblock_phys_alloc_xx() API.
- * The freeing memory will not be released to the buddy allocator.
+ * If called after the buddy allocator is available, the memory is released to
+ * the buddy allocator.
*/
int __init_memblock memblock_phys_free(phys_addr_t base, phys_addr_t size)
{
phys_addr_t end = base + size - 1;
+ int ret = 0;
memblock_dbg("%s: [%pa-%pa] %pS\n", __func__,
&base, &end, (void *)_RET_IP_);
kmemleak_free_part_phys(base, size);
- return memblock_remove_range(&memblock.reserved, base, size);
+
+ if (!slab_is_available() || IS_ENABLED(CONFIG_ARCH_KEEP_MEMBLOCK))
+ ret = memblock_remove_range(&memblock.reserved, base, size);
+
+ if (slab_is_available())
+ __free_reserved_area(base, base + size, -1);
+
+ return ret;
}
int __init_memblock __memblock_reserve(phys_addr_t base, phys_addr_t size,
@@ -949,40 +1027,6 @@ int __init_memblock memblock_physmem_add(phys_addr_t base, phys_addr_t size)
}
#endif
-#ifdef CONFIG_MEMBLOCK_KHO_SCRATCH
-__init void memblock_set_kho_scratch_only(void)
-{
- kho_scratch_only = true;
-}
-
-__init void memblock_clear_kho_scratch_only(void)
-{
- kho_scratch_only = false;
-}
-
-__init void memmap_init_kho_scratch_pages(void)
-{
- phys_addr_t start, end;
- unsigned long pfn;
- int nid;
- u64 i;
-
- if (!IS_ENABLED(CONFIG_DEFERRED_STRUCT_PAGE_INIT))
- return;
-
- /*
- * Initialize struct pages for free scratch memory.
- * The struct pages for reserved scratch memory will be set up in
- * reserve_bootmem_region()
- */
- __for_each_mem_range(i, &memblock.memory, NULL, NUMA_NO_NODE,
- MEMBLOCK_KHO_SCRATCH, &start, &end, &nid) {
- for (pfn = PFN_UP(start); pfn < PFN_DOWN(end); pfn++)
- init_deferred_page(pfn, nid);
- }
-}
-#endif
-
/**
* memblock_setclr_flag - set or clear flag for a memory region
* @type: memblock type to set/clear flag for
@@ -1116,6 +1160,21 @@ int __init_memblock memblock_reserved_mark_noinit(phys_addr_t base, phys_addr_t
}
/**
+ * memblock_reserved_mark_kern - Mark a reserved memory region with flag
+ * MEMBLOCK_RSRV_KERN
+ *
+ * @base: the base phys addr of the region
+ * @size: the size of the region
+ *
+ * Return: 0 on success, -errno on failure.
+ */
+int __init_memblock memblock_reserved_mark_kern(phys_addr_t base, phys_addr_t size)
+{
+ return memblock_setclr_flag(&memblock.reserved, base, size, 1,
+ MEMBLOCK_RSRV_KERN);
+}
+
+/**
* memblock_mark_kho_scratch - Mark a memory region as MEMBLOCK_KHO_SCRATCH.
* @base: the base phys addr of the region
* @size: the size of the region
@@ -1446,6 +1505,32 @@ int __init_memblock memblock_set_node(phys_addr_t base, phys_addr_t size,
return 0;
}
+static void memblock_prep_allocation(phys_addr_t start, phys_addr_t size,
+ bool kmemleak_trace)
+{
+ /*
+ * Skip kmemleak for those places like kasan_init() and
+ * early_pgtable_alloc() due to high volume.
+ */
+ if (kmemleak_trace)
+ /*
+ * Memblock allocated blocks are never reported as
+ * leaks. This is because many of these blocks are
+ * only referred via the physical address which is
+ * not looked up by kmemleak.
+ */
+ kmemleak_alloc_phys(start, size, 0);
+
+ /*
+ * Some Virtual Machine platforms, such as Intel TDX or AMD SEV-SNP,
+ * require memory to be accepted before it can be used by the
+ * guest.
+ *
+ * Accept the memory of the allocated buffer.
+ */
+ accept_memory(start, size);
+}
+
/**
* memblock_alloc_range_nid - allocate boot memory block
* @size: size of memory block to be allocated in bytes
@@ -1520,28 +1605,7 @@ again:
return 0;
done:
- /*
- * Skip kmemleak for those places like kasan_init() and
- * early_pgtable_alloc() due to high volume.
- */
- if (end != MEMBLOCK_ALLOC_NOLEAKTRACE)
- /*
- * Memblock allocated blocks are never reported as
- * leaks. This is because many of these blocks are
- * only referred via the physical address which is
- * not looked up by kmemleak.
- */
- kmemleak_alloc_phys(found, size, 0);
-
- /*
- * Some Virtual Machine platforms, such as Intel TDX or AMD SEV-SNP,
- * require memory to be accepted before it can be used by the
- * guest.
- *
- * Accept the memory of the allocated buffer.
- */
- accept_memory(found, size);
-
+ memblock_prep_allocation(found, size, end != MEMBLOCK_ALLOC_NOLEAKTRACE);
return found;
}
@@ -1697,6 +1761,77 @@ void * __init memblock_alloc_try_nid_raw(
}
/**
+ * memblock_alloc_hugetlb - allocate boot memory for HugeTLB pages
+ * @size: size of the memory to be allocated in bytes
+ * @nid: nid of the free memory to find, %NUMA_NO_NODE for any node
+ * @exact_nid: only allocate from the specified nid. If %false, the specified
+ * nid is tried first, and then all nodes are tried as fallback.
+ *
+ * HugeTLB pages are always aligned by their size, so the alignment matches
+ * @size. Since the memory is for userspace, mirrored memory is not used. The
+ * memory is not zeroed. Does not panic if request cannot be satisfied.
+ *
+ * Return:
+ * Virtual address of allocated memory block on success, %NULL on failure.
+ */
+void * __init memblock_alloc_hugetlb(phys_addr_t size, int nid, bool exact_nid)
+{
+ enum memblock_flags flags = choose_memblock_flags();
+ phys_addr_t addr, start = 0, end = MEMBLOCK_ALLOC_ACCESSIBLE;
+
+ memblock_dbg("%s: %llu bytes, nid=%d, exact_nid=%d %pS\n", __func__,
+ (u64)size, nid, exact_nid, (void *)_RET_IP_);
+
+ /* Don't waste mirrored memory on HugeTLB pages. */
+ flags &= ~MEMBLOCK_MIRROR;
+retry:
+ /* HugeTLB pages are always aligned by their size. */
+ addr = memblock_find_in_range_node(size, size, start, end, nid, flags);
+ if (addr)
+ goto found;
+
+ /* Try all nodes if allowed. */
+ if (numa_valid_node(nid) && !exact_nid) {
+ nid = NUMA_NO_NODE;
+ /*
+ * If a previous candidate overlapped with KHO scratch, it would
+ * update start or end. Now that the search is opening to all
+ * nodes, reset them.
+ */
+ start = 0;
+ end = MEMBLOCK_ALLOC_ACCESSIBLE;
+
+ goto retry;
+ }
+
+ /* Found nothing... :-( */
+ return NULL;
+
+found:
+ /*
+ * HugeTLB pages can be preserved with KHO and no preserved memory can
+ * be in scratch. So retry if found address overlaps with scratch.
+ *
+ * Scratch areas are normally not very large, so this shouldn't take too
+ * many retries.
+ */
+ if (kho_scratch_overlap(addr, size)) {
+ if (memblock_bottom_up())
+ start = addr + size;
+ else
+ end = addr;
+
+ goto retry;
+ }
+
+ if (__memblock_reserve(addr, size, nid, MEMBLOCK_RSRV_KERN | MEMBLOCK_RSRV_HUGETLB))
+ return NULL;
+
+ memblock_prep_allocation(addr, size, true);
+ return phys_to_virt(addr);
+}
+
+/**
* memblock_alloc_try_nid - allocate boot memory block
* @size: size of memory block to be allocated in bytes
* @align: alignment of the region and block's size
@@ -1751,32 +1886,6 @@ void *__init __memblock_alloc_or_panic(phys_addr_t size, phys_addr_t align,
return addr;
}
-/**
- * memblock_free_late - free pages directly to buddy allocator
- * @base: phys starting address of the boot memory block
- * @size: size of the boot memory block in bytes
- *
- * This is only useful when the memblock allocator has already been torn
- * down, but we are still initializing the system. Pages are released directly
- * to the buddy allocator.
- */
-void __init memblock_free_late(phys_addr_t base, phys_addr_t size)
-{
- phys_addr_t cursor, end;
-
- end = base + size - 1;
- memblock_dbg("%s: [%pa-%pa] %pS\n",
- __func__, &base, &end, (void *)_RET_IP_);
- kmemleak_free_part_phys(base, size);
- cursor = PFN_UP(base);
- end = PFN_DOWN(base + size);
-
- for (; cursor < end; cursor++) {
- memblock_free_pages(cursor, 0);
- totalram_pages_inc();
- }
-}
-
/*
* Remaining API functions
*/
@@ -1791,6 +1900,28 @@ phys_addr_t __init_memblock memblock_reserved_size(void)
return memblock.reserved.total_size;
}
+phys_addr_t __init_memblock memblock_reserved_hugetlb_size(phys_addr_t limit, int nid)
+{
+ struct memblock_region *r;
+ phys_addr_t total = 0;
+
+ for_each_reserved_mem_region(r) {
+ phys_addr_t size = r->size;
+
+ if (r->base > limit)
+ break;
+
+ if (r->base + r->size > limit)
+ size = limit - r->base;
+
+ if (nid == memblock_get_region_node(r) || !numa_valid_node(nid))
+ if (r->flags & MEMBLOCK_RSRV_HUGETLB)
+ total += size;
+ }
+
+ return total;
+}
+
phys_addr_t __init_memblock memblock_reserved_kern_size(phys_addr_t limit, int nid)
{
struct memblock_region *r;
@@ -2190,10 +2321,8 @@ static void __init free_unused_memmap(void)
}
#ifdef CONFIG_SPARSEMEM
- if (!IS_ALIGNED(prev_end, PAGES_PER_SECTION)) {
- prev_end = pageblock_align(end);
+ if (!IS_ALIGNED(prev_end, PAGES_PER_SECTION))
free_memmap(prev_end, ALIGN(prev_end, PAGES_PER_SECTION));
- }
#endif
}
@@ -2240,6 +2369,31 @@ static unsigned long __init __free_memory_core(phys_addr_t start,
return end_pfn - start_pfn;
}
+/*
+ * Initialised pages do not have PageReserved set. This function is called
+ * for each reserved range and marks the pages PageReserved.
+ * When deferred initialization of struct pages is enabled it also ensures
+ * that struct pages are properly initialised.
+ */
+static void __init memmap_init_reserved_range(phys_addr_t start,
+ phys_addr_t end, int nid)
+{
+ unsigned long pfn;
+
+ for_each_valid_pfn(pfn, PFN_DOWN(start), PFN_UP(end)) {
+ struct page *page = pfn_to_page(pfn);
+
+ init_deferred_page(pfn, nid);
+
+ /*
+ * no need for atomic set_bit because the struct
+ * page is not visible yet so nobody should
+ * access it yet.
+ */
+ __SetPageReserved(page);
+ }
+}
+
static void __init memmap_init_reserved_pages(void)
{
struct memblock_region *region;
@@ -2259,7 +2413,7 @@ repeat:
end = start + region->size;
if (memblock_is_nomap(region))
- reserve_bootmem_region(start, end, nid);
+ memmap_init_reserved_range(start, end, nid);
memblock_set_node(start, region->size, &memblock.reserved, nid);
}
@@ -2284,7 +2438,7 @@ repeat:
if (!numa_valid_node(nid))
nid = early_pfn_to_nid(PFN_DOWN(start));
- reserve_bootmem_region(start, end, nid);
+ memmap_init_reserved_range(start, end, nid);
}
}
}
@@ -2434,7 +2588,7 @@ int reserve_mem_release_by_name(const char *name)
return 0;
start = phys_to_virt(map->start);
- end = start + map->size - 1;
+ end = start + map->size;
snprintf(buf, sizeof(buf), "reserve_mem:%s", name);
free_reserved_area(start, end, 0, buf);
map->size = 0;
@@ -2442,6 +2596,18 @@ int reserve_mem_release_by_name(const char *name)
return 1;
}
+#ifdef CONFIG_MEMBLOCK_KHO_SCRATCH
+__init void memblock_set_kho_scratch_only(void)
+{
+ kho_scratch_only = true;
+}
+
+__init void memblock_clear_kho_scratch_only(void)
+{
+ kho_scratch_only = false;
+}
+#endif
+
#ifdef CONFIG_KEXEC_HANDOVER
static int __init reserved_mem_preserve(void)
@@ -2510,7 +2676,7 @@ static int __init prepare_kho_fdt(void)
if (err)
goto err_unpreserve_fdt;
- err = kho_add_subtree(MEMBLOCK_KHO_FDT, fdt);
+ err = kho_add_subtree(MEMBLOCK_KHO_FDT, fdt, fdt_totalsize(fdt));
if (err)
goto err_unpreserve_fdt;
@@ -2555,7 +2721,7 @@ static void *__init reserve_mem_kho_retrieve_fdt(void)
if (fdt)
return fdt;
- err = kho_retrieve_subtree(MEMBLOCK_KHO_FDT, &fdt_phys);
+ err = kho_retrieve_subtree(MEMBLOCK_KHO_FDT, &fdt_phys, NULL);
if (err) {
if (err != -ENOENT)
pr_warn("failed to retrieve FDT '%s' from KHO: %d\n",
@@ -2642,23 +2808,25 @@ static int __init reserve_mem(char *p)
int len;
if (!p)
- return -EINVAL;
+ goto err_param;
/* Check if there's room for more reserved memory */
- if (reserved_mem_count >= RESERVE_MEM_MAX_ENTRIES)
+ if (reserved_mem_count >= RESERVE_MEM_MAX_ENTRIES) {
+ pr_err("reserve_mem: no more room for reserved memory\n");
return -EBUSY;
+ }
oldp = p;
size = memparse(p, &p);
if (!size || p == oldp)
- return -EINVAL;
+ goto err_param;
if (*p != ':')
- return -EINVAL;
+ goto err_param;
align = memparse(p+1, &p);
if (*p != ':')
- return -EINVAL;
+ goto err_param;
/*
* memblock_phys_alloc() doesn't like a zero size align,
@@ -2672,7 +2840,7 @@ static int __init reserve_mem(char *p)
/* name needs to have length but not too big */
if (!len || len >= RESERVE_MEM_NAME_SIZE)
- return -EINVAL;
+ goto err_param;
/* Make sure that name has text */
for (p = name; *p; p++) {
@@ -2680,11 +2848,13 @@ static int __init reserve_mem(char *p)
break;
}
if (!*p)
- return -EINVAL;
+ goto err_param;
/* Make sure the name is not already used */
- if (reserve_mem_find_by_name(name, &start, &tmp))
+ if (reserve_mem_find_by_name(name, &start, &tmp)) {
+ pr_err("reserve_mem: name \"%s\" was already used\n", name);
return -EBUSY;
+ }
/* Pick previous allocations up from KHO if available */
if (reserve_mem_kho_revive(name, size, align))
@@ -2692,16 +2862,22 @@ static int __init reserve_mem(char *p)
/* TODO: Allocation must be outside of scratch region */
start = memblock_phys_alloc(size, align);
- if (!start)
+ if (!start) {
+ pr_err("reserve_mem: memblock allocation failed\n");
return -ENOMEM;
+ }
reserved_mem_add(start, size, name);
return 1;
+err_param:
+ pr_err("reserve_mem: empty or malformed parameter\n");
+ return -EINVAL;
}
__setup("reserve_mem=", reserve_mem);
-#if defined(CONFIG_DEBUG_FS) && defined(CONFIG_ARCH_KEEP_MEMBLOCK)
+#ifdef CONFIG_DEBUG_FS
+#ifdef CONFIG_ARCH_KEEP_MEMBLOCK
static const char * const flagname[] = {
[ilog2(MEMBLOCK_HOTPLUG)] = "HOTPLUG",
[ilog2(MEMBLOCK_MIRROR)] = "MIRROR",
@@ -2732,14 +2908,18 @@ static int memblock_debug_show(struct seq_file *m, void *private)
else
seq_printf(m, "%4c ", 'x');
if (reg->flags) {
- for (j = 0; j < count; j++) {
- if (reg->flags & (1U << j)) {
- seq_printf(m, "%s\n", flagname[j]);
- break;
- }
+ unsigned int flags = reg->flags;
+ bool first = true;
+
+ for (j = 0; flags; j++, flags >>= 1) {
+ if (!(flags & 1))
+ continue;
+ if (!first)
+ seq_putc(m, '|');
+ seq_puts(m, j < count ? flagname[j] : "UNKNOWN");
+ first = false;
}
- if (j == count)
- seq_printf(m, "%s\n", "UNKNOWN");
+ seq_putc(m, '\n');
} else {
seq_printf(m, "%s\n", "NONE");
}
@@ -2748,10 +2928,8 @@ static int memblock_debug_show(struct seq_file *m, void *private)
}
DEFINE_SHOW_ATTRIBUTE(memblock_debug);
-static int __init memblock_init_debugfs(void)
+static inline void memblock_debugfs_expose_arrays(struct dentry *root)
{
- struct dentry *root = debugfs_create_dir("memblock", NULL);
-
debugfs_create_file("memory", 0444, root,
&memblock.memory, &memblock_debug_fops);
debugfs_create_file("reserved", 0444, root,
@@ -2760,7 +2938,48 @@ static int __init memblock_init_debugfs(void)
debugfs_create_file("physmem", 0444, root, &physmem,
&memblock_debug_fops);
#endif
+}
+
+#else
+
+static inline void memblock_debugfs_expose_arrays(struct dentry *root) { }
+
+#endif /* CONFIG_ARCH_KEEP_MEMBLOCK */
+
+static int memblock_reserve_mem_show(struct seq_file *m, void *private)
+{
+ struct reserve_mem_table *map;
+ char txtsz[16];
+
+ guard(mutex)(&reserve_mem_lock);
+ for (int i = 0; i < reserved_mem_count; i++) {
+ map = &reserved_mem_table[i];
+ if (!map->size)
+ continue;
+
+ memset(txtsz, 0, sizeof(txtsz));
+ string_get_size(map->size, 1, STRING_UNITS_2, txtsz, sizeof(txtsz));
+ seq_printf(m, "%s\t\t(%s)\n", map->name, txtsz);
+ }
+
+ return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(memblock_reserve_mem);
+
+static int __init memblock_init_debugfs(void)
+{
+ struct dentry *root;
+
+ if (!IS_ENABLED(CONFIG_ARCH_KEEP_MEMBLOCK) && !reserved_mem_count)
+ return 0;
+
+ root = debugfs_create_dir("memblock", NULL);
+
+ if (reserved_mem_count)
+ debugfs_create_file("reserve_mem_param", 0444, root, NULL,
+ &memblock_reserve_mem_fops);
+ memblock_debugfs_expose_arrays(root);
return 0;
}
__initcall(memblock_init_debugfs);