Revert "mm: Always release pages to the buddy allocator in memblock_free_late()."
authorAaron Thompson <dev@aaront.org>
Tue, 7 Feb 2023 08:21:51 +0000 (08:21 +0000)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 22 Feb 2023 11:59:50 +0000 (12:59 +0100)
commit 647037adcad00f2bab8828d3d41cd0553d41f3bd upstream.

This reverts commit 115d9d77bb0f9152c60b6e8646369fa7f6167593.

The pages being freed by memblock_free_late() have already been
initialized, but if they are in the deferred init range,
__free_one_page() might access nearby uninitialized pages when trying to
coalesce buddies. This can, for example, trigger this BUG:

  BUG: unable to handle page fault for address: ffffe964c02580c8
  RIP: 0010:__list_del_entry_valid+0x3f/0x70
   <TASK>
   __free_one_page+0x139/0x410
   __free_pages_ok+0x21d/0x450
   memblock_free_late+0x8c/0xb9
   efi_free_boot_services+0x16b/0x25c
   efi_enter_virtual_mode+0x403/0x446
   start_kernel+0x678/0x714
   secondary_startup_64_no_verify+0xd2/0xdb
   </TASK>

A proper fix will be more involved so revert this change for the time
being.

Fixes: 115d9d77bb0f ("mm: Always release pages to the buddy allocator in memblock_free_late().")
Signed-off-by: Aaron Thompson <dev@aaront.org>
Link: https://lore.kernel.org/r/20230207082151.1303-1-dev@aaront.org
Signed-off-by: Mike Rapoport (IBM) <rppt@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mm/memblock.c
tools/testing/memblock/internal.h

index fc3d8fb..511d478 100644 (file)
@@ -1640,13 +1640,7 @@ void __init memblock_free_late(phys_addr_t base, phys_addr_t size)
        end = PFN_DOWN(base + size);
 
        for (; cursor < end; cursor++) {
-               /*
-                * Reserved pages are always initialized by the end of
-                * memblock_free_all() (by memmap_init() and, if deferred
-                * initialization is enabled, memmap_init_reserved_pages()), so
-                * these pages can be released directly to the buddy allocator.
-                */
-               __free_pages_core(pfn_to_page(cursor), 0);
+               memblock_free_pages(pfn_to_page(cursor), cursor, 0);
                totalram_pages_inc();
        }
 }
index 85973e5..fdb7f5d 100644 (file)
@@ -15,10 +15,6 @@ bool mirrored_kernelcore = false;
 
 struct page {};
 
-void __free_pages_core(struct page *page, unsigned int order)
-{
-}
-
 void memblock_free_pages(struct page *page, unsigned long pfn,
                         unsigned int order)
 {