arm64: memblock: don't permit memblock resizing until linear mapping is up

Bhupesh reports that having numerous memblock reservations at early
boot may result in the following crash:

  Unable to handle kernel paging request at virtual address ffff80003ffe0000
  ...
  Call trace:
   __memcpy+0x110/0x180
   memblock_add_range+0x134/0x2e8
   memblock_reserve+0x70/0xb8
   memblock_alloc_base_nid+0x6c/0x88
   __memblock_alloc_base+0x3c/0x4c
   memblock_alloc_base+0x28/0x4c
   memblock_alloc+0x2c/0x38
   early_pgtable_alloc+0x20/0xb0
   paging_init+0x28/0x7f8

This is caused by the fact that we permit memblock resizing before the
linear mapping is up, and so the memblock_reserved() array is moved
into memory that is not mapped yet.

So let's ensure that this crash can no longer occur, by deferring to
call to memblock_allow_resize() to after the linear mapping has been
created.

Reported-by: Bhupesh Sharma <[email protected]>
Acked-by: Will Deacon <[email protected]>
Tested-by: Marc Zyngier <[email protected]>
Signed-off-by: Ard Biesheuvel <[email protected]>
Signed-off-by: Catalin Marinas <[email protected]>

Author: Ard Biesheuvel

arch/arm64/mm/init.c

@@ -483,8 +483,6 @@ void __init arm64_memblock_init(void)
 	high_memory = __va(memblock_end_of_DRAM() - 1) + 1;
 
 	dma_contiguous_reserve(arm64_dma_phys_limit);
-
-	memblock_allow_resize();
 }
 
 void __init bootmem_init(void)

arch/arm64/mm/mmu.c

@@ -659,6 +659,8 @@ void __init paging_init(void)
 
 	memblock_free(__pa_symbol(init_pg_dir),
 		      __pa_symbol(init_pg_end) - __pa_symbol(init_pg_dir));
+
+	memblock_allow_resize();
 }
 
 /*