mm/mm_init: don't overlap zones with kernelcore=mirror

From: "Mike Rapoport (Microsoft)" <[email protected]>

Hi,

These patches make the behaviour of kernelcore= parameter uniform and treat mirror just as another way to size the zones and cleanup a weird part of the memory map initialization.

For example, for the memory layout below with the first two memory ranges being mirrored (flags=0x2)

  memory[0x0]     [0x0000000000001000-0x000000000009efff], 0x000000000009e000 bytes on node 0 flags: 0x2
  memory[0x1]     [0x0000000000100000-0x00000000bffdefff], 0x00000000bfedf000 bytes on node 0 flags: 0x2
  memory[0x2]     [0x0000000100000000-0x000000013fffffff], 0x0000000040000000 bytes on node 0 flags: 0x2
  memory[0x3]     [0x0000000140000000-0x00000001bfffffff], 0x0000000080000000 bytes on node 0 flags: 0x0

with kernelcore=mirror set zone ranges would be

Normal [100000, 1c0000]
Movable [140000, 1c0000]

and range [140000, 1c0000] is spanned by both NORMAL and MOVABLE zones.

This range will be passed twice to memmap_init_range() - once for each zone that spans it.
The memory map for this range will be initialized as ZONE_NORMAL during the first pass and skipped because of overlap_memmap_init() during the second pass (ZONE_MOVABLE initialization), although the pages in this range actually belong to ZONE_MOVABLE.

Aligning kernelcore=mirror behaviour with other variants of kernelcore=/movablecore= resolves this issue and makes the code less obfuscated.

I intend to carry this via memblock tree.

Mike Rapoport (Microsoft) (2):
mm/mm_init: don't overlap NORMAL and MOVABLE zones with kernelcore=mirror mm/mm_init: drop overlap_memmap_init()

 mm/mm_init.c | 60 +++-------------------------------------------------
 1 file changed, 3 insertions(+), 57 deletions(-)

base-commit: dc59e4fea9d83f03bad6bddf3fa2e52491777482

2.53.0