Revert "x86/e820: put !E820_TYPE_RAM regions into memblock.reserved"
authorMasayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Fri, 26 Oct 2018 22:10:24 +0000 (15:10 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 26 Oct 2018 23:38:15 +0000 (16:38 -0700)
commit 124049decbb1 ("x86/e820: put !E820_TYPE_RAM regions into
memblock.reserved") breaks movable_node kernel option because it changed
the memory gap range to reserved memblock.  So, the node is marked as
Normal zone even if the SRAT has Hot pluggable affinity.

    =====================================================================
    kernel: BIOS-e820: [mem 0x0000180000000000-0x0000180fffffffff] usable
    kernel: BIOS-e820: [mem 0x00001c0000000000-0x00001c0fffffffff] usable
    ...
    kernel: reserved[0x12]#011[0x0000181000000000-0x00001bffffffffff], 0x000003f000000000 bytes flags: 0x0
    ...
    kernel: ACPI: SRAT: Node 2 PXM 6 [mem 0x180000000000-0x1bffffffffff] hotplug
    kernel: ACPI: SRAT: Node 3 PXM 7 [mem 0x1c0000000000-0x1fffffffffff] hotplug
    ...
    kernel: Movable zone start for each node
    kernel:  Node 3: 0x00001c0000000000
    kernel: Early memory node ranges
    ...
    =====================================================================

The original issue is fixed by the former patches, so let's revert commit
124049decbb1 ("x86/e820: put !E820_TYPE_RAM regions into
memblock.reserved").

Link: http://lkml.kernel.org/r/20181002143821.5112-4-msys.mizuma@gmail.com
Signed-off-by: Masayoshi Mizuma <m.mizuma@jp.fujitsu.com>
Reviewed-by: Pavel Tatashin <pavel.tatashin@microsoft.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Oscar Salvador <osalvador@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
arch/x86/kernel/e820.c

index c88c23c658c1e99faad3daa236448bc4208901d7..d1f25c83144752272401afe8c8aec313d40298ae 100644 (file)
@@ -1248,7 +1248,6 @@ void __init e820__memblock_setup(void)
 {
        int i;
        u64 end;
-       u64 addr = 0;
 
        /*
         * The bootstrap memblock region count maximum is 128 entries
@@ -1265,21 +1264,13 @@ void __init e820__memblock_setup(void)
                struct e820_entry *entry = &e820_table->entries[i];
 
                end = entry->addr + entry->size;
-               if (addr < entry->addr)
-                       memblock_reserve(addr, entry->addr - addr);
-               addr = end;
                if (end != (resource_size_t)end)
                        continue;
 
-               /*
-                * all !E820_TYPE_RAM ranges (including gap ranges) are put
-                * into memblock.reserved to make sure that struct pages in
-                * such regions are not left uninitialized after bootup.
-                */
                if (entry->type != E820_TYPE_RAM && entry->type != E820_TYPE_RESERVED_KERN)
-                       memblock_reserve(entry->addr, entry->size);
-               else
-                       memblock_add(entry->addr, entry->size);
+                       continue;
+
+               memblock_add(entry->addr, entry->size);
        }
 
        /* Throw away partial pages: */