[PATCH v4 1/2] kho: fix deferred init of kho scratch

[Mike Rapoport]

On Fri, Feb 20, 2026 at 05:52:02PM +0100, Michal Clapinski wrote:
> Currently, mm_core_init calls kho_memory_init, which calls
> kho_release_scratch.
> 
> If DEFERRED is enabled, kho_release_scratch will first initialize the
> struct pages of kho scratch. This is not needed. We can just let
> page_alloc_init_late init it.
> 
> Next, kho_release_scratch will mark scratch as MIGRATE_CMA. If DEFERRED
> is enabled, this will be overwritten later in deferred_free_pages.
> 
> To fix this, I removed the whole kho_release_scratch.
> Marking the pageblocks as MIGRATE_CMA now happens in kho_init, which
> runs after deferred_free_pages.
> 
> Signed-off-by: Michal Clapinski <mclapinski at google.com>

Reviewed-by: Mike Rapoport (Microsoft) <rppt at kernel.org>

> ---
>  include/linux/memblock.h           |  2 --
>  kernel/liveupdate/kexec_handover.c | 43 ++++++++----------------------
>  mm/memblock.c                      | 22 ---------------
>  3 files changed, 11 insertions(+), 56 deletions(-)
> 
> diff --git a/include/linux/memblock.h b/include/linux/memblock.h
> index 221118b5a16e..35d9cf6bbf7a 100644
> --- a/include/linux/memblock.h
> +++ b/include/linux/memblock.h
> @@ -614,11 +614,9 @@ static inline void memtest_report_meminfo(struct seq_file *m) { }
>  #ifdef CONFIG_MEMBLOCK_KHO_SCRATCH
>  void memblock_set_kho_scratch_only(void);
>  void memblock_clear_kho_scratch_only(void);
> -void memmap_init_kho_scratch_pages(void);
>  #else
>  static inline void memblock_set_kho_scratch_only(void) { }
>  static inline void memblock_clear_kho_scratch_only(void) { }
> -static inline void memmap_init_kho_scratch_pages(void) {}
>  #endif
>  
>  #endif /* _LINUX_MEMBLOCK_H */
> diff --git a/kernel/liveupdate/kexec_handover.c b/kernel/liveupdate/kexec_handover.c
> index b851b09a8e99..de167bfa2c8d 100644
> --- a/kernel/liveupdate/kexec_handover.c
> +++ b/kernel/liveupdate/kexec_handover.c
> @@ -1377,11 +1377,6 @@ static __init int kho_init(void)
>  	if (err)
>  		goto err_free_fdt;
>  
> -	if (fdt) {
> -		kho_in_debugfs_init(&kho_in.dbg, fdt);
> -		return 0;
> -	}
> -
>  	for (int i = 0; i < kho_scratch_cnt; i++) {
>  		unsigned long base_pfn = PHYS_PFN(kho_scratch[i].addr);
>  		unsigned long count = kho_scratch[i].size >> PAGE_SHIFT;
> @@ -1397,8 +1392,17 @@ static __init int kho_init(void)
>  		 */
>  		kmemleak_ignore_phys(kho_scratch[i].addr);
>  		for (pfn = base_pfn; pfn < base_pfn + count;
> -		     pfn += pageblock_nr_pages)
> -			init_cma_reserved_pageblock(pfn_to_page(pfn));
> +		     pfn += pageblock_nr_pages) {
> +			if (fdt)
> +				init_cma_pageblock(pfn_to_page(pfn));
> +			else
> +				init_cma_reserved_pageblock(pfn_to_page(pfn));
> +		}
> +	}
> +
> +	if (fdt) {
> +		kho_in_debugfs_init(&kho_in.dbg, fdt);
> +		return 0;
>  	}
>  
>  	WARN_ON_ONCE(kho_debugfs_fdt_add(&kho_out.dbg, "fdt",
> @@ -1421,35 +1425,10 @@ static __init int kho_init(void)
>  }
>  fs_initcall(kho_init);
>  
> -static void __init kho_release_scratch(void)
> -{
> -	phys_addr_t start, end;
> -	u64 i;
> -
> -	memmap_init_kho_scratch_pages();
> -
> -	/*
> -	 * Mark scratch mem as CMA before we return it. That way we
> -	 * ensure that no kernel allocations happen on it. That means
> -	 * we can reuse it as scratch memory again later.
> -	 */
> -	__for_each_mem_range(i, &memblock.memory, NULL, NUMA_NO_NODE,
> -			     MEMBLOCK_KHO_SCRATCH, &start, &end, NULL) {
> -		ulong start_pfn = pageblock_start_pfn(PFN_DOWN(start));
> -		ulong end_pfn = pageblock_align(PFN_UP(end));
> -		ulong pfn;
> -
> -		for (pfn = start_pfn; pfn < end_pfn; pfn += pageblock_nr_pages)
> -			init_pageblock_migratetype(pfn_to_page(pfn),
> -						   MIGRATE_CMA, false);
> -	}
> -}
> -
>  void __init kho_memory_init(void)
>  {
>  	if (kho_in.mem_map_phys) {
>  		kho_scratch = phys_to_virt(kho_in.scratch_phys);
> -		kho_release_scratch();
>  		kho_mem_deserialize(phys_to_virt(kho_in.mem_map_phys));
>  	} else {
>  		kho_reserve_scratch();
> diff --git a/mm/memblock.c b/mm/memblock.c
> index 6cff515d82f4..3eff19124fc0 100644
> --- a/mm/memblock.c
> +++ b/mm/memblock.c
> @@ -959,28 +959,6 @@ __init void memblock_clear_kho_scratch_only(void)
>  {
>  	kho_scratch_only = false;
>  }
> -
> -__init void memmap_init_kho_scratch_pages(void)
> -{
> -	phys_addr_t start, end;
> -	unsigned long pfn;
> -	int nid;
> -	u64 i;
> -
> -	if (!IS_ENABLED(CONFIG_DEFERRED_STRUCT_PAGE_INIT))
> -		return;
> -
> -	/*
> -	 * Initialize struct pages for free scratch memory.
> -	 * The struct pages for reserved scratch memory will be set up in
> -	 * reserve_bootmem_region()
> -	 */
> -	__for_each_mem_range(i, &memblock.memory, NULL, NUMA_NO_NODE,
> -			     MEMBLOCK_KHO_SCRATCH, &start, &end, &nid) {
> -		for (pfn = PFN_UP(start); pfn < PFN_DOWN(end); pfn++)
> -			init_deferred_page(pfn, nid);
> -	}
> -}
>  #endif
>  
>  /**
> -- 
> 2.53.0.345.g96ddfc5eaa-goog
> 

-- 
Sincerely yours,
Mike.