[PATCH 1/4] mm/sparse: nr_pages won't be 0

[PATCH 1/4] mm/sparse: nr_pages won't be 0

[Wei Yang]

Function subsection_map_init() is only used in free_area_init() in the
loop of for_each_mem_pfn_range(). And we are sure in each iteration of
for_each_mem_pfn_range(), start_pfn < end_pfn.

So nr_pages is not possible to be 0 and we can remove the check.

Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
CC: Mike Rapoport (IBM) <rppt@kernel.org>
---
 mm/sparse.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/mm/sparse.c b/mm/sparse.c
index de40b2c73406..f56701873ce8 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -195,9 +195,6 @@ void __init subsection_map_init(unsigned long pfn, unsigned long nr_pages)
 	int end_sec = pfn_to_section_nr(pfn + nr_pages - 1);
 	unsigned long nr, start_sec = pfn_to_section_nr(pfn);
 
-	if (!nr_pages)
-		return;
-
 	for (nr = start_sec; nr <= end_sec; nr++) {
 		struct mem_section *ms;
 		unsigned long pfns;
-- 
2.34.1

[PATCH 2/4] mm/mm_init.c: move build check on MAX_ZONELISTS out of ifdef

[Wei Yang]

Current check on MAX_ZONELISTS is wrapped in CONFIG_DEBUG_MEMORY_INIT,
which may not be triggered all the time.

Let's move it out to a more general place.

Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
CC: Mike Rapoport (IBM) <rppt@kernel.org>
---
 mm/mm_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/mm_init.c b/mm/mm_init.c
index 31ebb3cd53ac..4991cb5c5436 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -53,7 +53,6 @@ void __init mminit_verify_zonelist(void)
 		struct zonelist *zonelist;
 		int i, listid, zoneid;
 
-		BUILD_BUG_ON(MAX_ZONELISTS > 2);
 		for (i = 0; i < MAX_ZONELISTS * MAX_NR_ZONES; i++) {
 
 			/* Identify the zone and nodelist */
@@ -2668,6 +2667,7 @@ static void __init mem_init_print_info(void)
 void __init mm_core_init(void)
 {
 	/* Initializations relying on SMP setup */
+	BUILD_BUG_ON(MAX_ZONELISTS > 2);
 	build_all_zonelists(NULL);
 	page_alloc_init_cpuhp();
 
-- 
2.34.1

Re: [PATCH 2/4] mm/mm_init.c: move build check on MAX_ZONELISTS out of ifdef

[Mike Rapoport]

On Wed, Jun 19, 2024 at 01:06:10AM +0000, Wei Yang wrote:
> Current check on MAX_ZONELISTS is wrapped in CONFIG_DEBUG_MEMORY_INIT,
> which may not be triggered all the time.
> 
> Let's move it out to a more general place.
> 
> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
> CC: Mike Rapoport (IBM) <rppt@kernel.org>

Reviewed-by: Mike Rapoport (IBM) <rppt@kernel.org>

> ---
>  mm/mm_init.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/mm_init.c b/mm/mm_init.c
> index 31ebb3cd53ac..4991cb5c5436 100644
> --- a/mm/mm_init.c
> +++ b/mm/mm_init.c
> @@ -53,7 +53,6 @@ void __init mminit_verify_zonelist(void)
>  		struct zonelist *zonelist;
>  		int i, listid, zoneid;
>  
> -		BUILD_BUG_ON(MAX_ZONELISTS > 2);
>  		for (i = 0; i < MAX_ZONELISTS * MAX_NR_ZONES; i++) {
>  
>  			/* Identify the zone and nodelist */
> @@ -2668,6 +2667,7 @@ static void __init mem_init_print_info(void)
>  void __init mm_core_init(void)
>  {
>  	/* Initializations relying on SMP setup */
> +	BUILD_BUG_ON(MAX_ZONELISTS > 2);
>  	build_all_zonelists(NULL);
>  	page_alloc_init_cpuhp();
>  
> -- 
> 2.34.1
> 

-- 
Sincerely yours,
Mike.

[PATCH 3/4] mm/page_alloc: fix a typo in comment about GFP flag

[Wei Yang]

The GFP flags used to choose the zonelist is __GFP_THISNODE.

Let's change it to what exactly it should be.

Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
---
 mm/page_alloc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f3ba1ede142d..594176449890 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -5209,7 +5209,7 @@ static void build_zonelists_in_node_order(pg_data_t *pgdat, int *node_order,
 }
 
 /*
- * Build gfp_thisnode zonelists
+ * Build __GFP_THISNODE zonelists
  */
 static void build_thisnode_zonelists(pg_data_t *pgdat)
 {
-- 
2.34.1

Re: [PATCH 3/4] mm/page_alloc: fix a typo in comment about GFP flag

[Mike Rapoport]

On Wed, Jun 19, 2024 at 01:06:11AM +0000, Wei Yang wrote:
> The GFP flags used to choose the zonelist is __GFP_THISNODE.
> 
> Let's change it to what exactly it should be.
> 
> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>

Acked-by: Mike Rapoport (IBM) <rppt@kernel.org>

> ---
>  mm/page_alloc.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index f3ba1ede142d..594176449890 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -5209,7 +5209,7 @@ static void build_zonelists_in_node_order(pg_data_t *pgdat, int *node_order,
>  }
>  
>  /*
> - * Build gfp_thisnode zonelists
> + * Build __GFP_THISNODE zonelists
>   */
>  static void build_thisnode_zonelists(pg_data_t *pgdat)
>  {
> -- 
> 2.34.1
> 

-- 
Sincerely yours,
Mike.

[PATCH 4/4] mm/page_alloc: reword the comment of buddy_merge_likely()

[Wei Yang]

For page with order O, we are checking its order (O + 1)'s buddy. If it
is free, we would like to put it to the tail and expect it would be
merged to a page with order (O + 2).

Reword the comment to reflect it.

Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
---
 mm/page_alloc.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 594176449890..d1df7929f003 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -706,12 +706,12 @@ static inline struct page *get_page_from_free_area(struct free_area *area,
 }
 
 /*
- * If this is not the largest possible page, check if the buddy
- * of the next-highest order is free. If it is, it's possible
+ * If this is less than the 2nd largest possible page, check if the buddy
+ * of the next-higher order is free. If it is, it's possible
  * that pages are being freed that will coalesce soon. In case,
  * that is happening, add the free page to the tail of the list
  * so it's less likely to be used soon and more likely to be merged
- * as a higher order page
+ * as a 2-level higher order page
  */
 static inline bool
 buddy_merge_likely(unsigned long pfn, unsigned long buddy_pfn,
-- 
2.34.1