memblock: free allocated memblock_reserved_regions later

memblock_free_reserved_regions() calls memblock_free(), but
memblock_free() would double reserved.regions too, so we could free the
old range for reserved.regions.

Also tj said there is another bug which could be related to this.

| I don't think we're saving any noticeable
| amount by doing this "free - give it to page allocator - reserve
| again" dancing.  We should just allocate regions aligned to page
| boundaries and free them later when memblock is no longer in use.

in that case, when DEBUG_PAGEALLOC, will get panic:

     memblock_free: [0x0000102febc080-0x0000102febf080] memblock_free_reserved_regions+0x37/0x39
  BUG: unable to handle kernel paging request at ffff88102febd948
  IP: [<ffffffff836a5774>] __next_free_mem_range+0x9b/0x155
  PGD 4826063 PUD cf67a067 PMD cf7fa067 PTE 800000102febd160
  Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC
  CPU 0
  Pid: 0, comm: swapper Not tainted 3.5.0-rc2-next-20120614-sasha #447
  RIP: 0010:[<ffffffff836a5774>]  [<ffffffff836a5774>] __next_free_mem_range+0x9b/0x155

See the discussion at https://lkml.org/lkml/2012/6/13/469

So try to allocate with PAGE_SIZE alignment and free it later.

Reported-by: Sasha Levin <levinsasha928@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

2 files changed, 46 insertions, 43 deletions

diff --git a/mm/memblock.c b/mm/memblock.c
index d4382095f8bd..5cc6731b00cc 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -143,30 +143,6 @@ phys_addr_t __init_memblock memblock_find_in_range(phys_addr_t start,
                                            MAX_NUMNODES);
 }
 
-/*
- * Free memblock.reserved.regions
- */
-int __init_memblock memblock_free_reserved_regions(void)
-{
-        if (memblock.reserved.regions == memblock_reserved_init_regions)
-                return 0;
-
-        return memblock_free(__pa(memblock.reserved.regions),
-                 sizeof(struct memblock_region) * memblock.reserved.max);
-}
-
-/*
- * Reserve memblock.reserved.regions
- */
-int __init_memblock memblock_reserve_reserved_regions(void)
-{
-        if (memblock.reserved.regions == memblock_reserved_init_regions)
-                return 0;
-
-        return memblock_reserve(__pa(memblock.reserved.regions),
-                 sizeof(struct memblock_region) * memblock.reserved.max);
-}
-
 static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r)
 {
         type->total_size -= type->regions[r].size;
@@ -184,6 +160,18 @@ static void __init_memblock memblock_remove_region(struct memblock_type *type, u
         }
 }
 
+phys_addr_t __init_memblock get_allocated_memblock_reserved_regions_info(
+                                        phys_addr_t *addr)
+{
+        if (memblock.reserved.regions == memblock_reserved_init_regions)
+                return 0;
+
+        *addr = __pa(memblock.reserved.regions);
+
+        return PAGE_ALIGN(sizeof(struct memblock_region) *
+                          memblock.reserved.max);
+}
+
 /**
  * memblock_double_array - double the size of the memblock regions array
  * @type: memblock type of the regions array being doubled
@@ -204,6 +192,7 @@ static int __init_memblock memblock_double_array(struct memblock_type *type,
                                                 phys_addr_t new_area_size)
 {
         struct memblock_region *new_array, *old_array;
+        phys_addr_t old_alloc_size, new_alloc_size;
         phys_addr_t old_size, new_size, addr;
         int use_slab = slab_is_available();
         int *in_slab;
@@ -217,6 +206,12 @@ static int __init_memblock memblock_double_array(struct memblock_type *type,
         /* Calculate new doubled size */
         old_size = type->max * sizeof(struct memblock_region);
         new_size = old_size << 1;
+        /*
+         * We need to allocated new one align to PAGE_SIZE,
+         *   so we can free them completely later.
+         */
+        old_alloc_size = PAGE_ALIGN(old_size);
+        new_alloc_size = PAGE_ALIGN(new_size);
 
         /* Retrieve the slab flag */
         if (type == &memblock.memory)
@@ -245,11 +240,11 @@ static int __init_memblock memblock_double_array(struct memblock_type *type,
 
                 addr = memblock_find_in_range(new_area_start + new_area_size,
                                                 memblock.current_limit,
-                                                new_size, sizeof(phys_addr_t));
+                                                new_alloc_size, PAGE_SIZE);
                 if (!addr && new_area_size)
                         addr = memblock_find_in_range(0,
                                         min(new_area_start, memblock.current_limit),
-                                        new_size, sizeof(phys_addr_t));
+                                        new_alloc_size, PAGE_SIZE);
 
                 new_array = addr ? __va(addr) : 0;
         }
@@ -279,13 +274,13 @@ static int __init_memblock memblock_double_array(struct memblock_type *type,
                 kfree(old_array);
         else if (old_array != memblock_memory_init_regions &&
                  old_array != memblock_reserved_init_regions)
-                memblock_free(__pa(old_array), old_size);
+                memblock_free(__pa(old_array), old_alloc_size);
 
         /* Reserve the new array if that comes from the memblock.
          * Otherwise, we needn't do it
          */
         if (!use_slab)
-                BUG_ON(memblock_reserve(addr, new_size));
+                BUG_ON(memblock_reserve(addr, new_alloc_size));
 
         /* Update slab flag */
         *in_slab = use_slab;
diff --git a/mm/nobootmem.c b/mm/nobootmem.c
index 0900b3910cda..405573010f99 100644
--- a/mm/nobootmem.c
+++ b/mm/nobootmem.c
@@ -105,27 +105,35 @@ static void __init __free_pages_memory(unsigned long start, unsigned long end)
                 __free_pages_bootmem(pfn_to_page(i), 0);
 }
 
+static unsigned long __init __free_memory_core(phys_addr_t start,
+                                 phys_addr_t end)
+{
+        unsigned long start_pfn = PFN_UP(start);
+        unsigned long end_pfn = min_t(unsigned long,
+                                      PFN_DOWN(end), max_low_pfn);
+
+        if (start_pfn > end_pfn)
+                return 0;
+
+        __free_pages_memory(start_pfn, end_pfn);
+
+        return end_pfn - start_pfn;
+}
+
 unsigned long __init free_low_memory_core_early(int nodeid)
 {
         unsigned long count = 0;
-        phys_addr_t start, end;
+        phys_addr_t start, end, size;
         u64 i;
 
-        /* free reserved array temporarily so that it's treated as free area */
-        memblock_free_reserved_regions();
-
-        for_each_free_mem_range(i, MAX_NUMNODES, &start, &end, NULL) {
-                unsigned long start_pfn = PFN_UP(start);
-                unsigned long end_pfn = min_t(unsigned long,
-                                              PFN_DOWN(end), max_low_pfn);
-                if (start_pfn < end_pfn) {
-                        __free_pages_memory(start_pfn, end_pfn);
-                        count += end_pfn - start_pfn;
-                }
-        }
+        for_each_free_mem_range(i, MAX_NUMNODES, &start, &end, NULL)
+                count += __free_memory_core(start, end);
+
+        /* free range that is used for reserved array if we allocate it */
+        size = get_allocated_memblock_reserved_regions_info(&start);
+        if (size)
+                count += __free_memory_core(start, start + size);
 
-        /* put region array back? */
-        memblock_reserve_reserved_regions();
         return count;
 }