diff mbox series

[v3,2/7] hugetlb: split hugetlb_hstate_alloc_pages

Message ID 20240102131249.76622-3-gang.li@linux.dev
State New
Headers
Series hugetlb: parallelize hugetlb page init on boot |

Commit Message

Gang Li Jan. 2, 2024, 1:12 p.m. UTC 
  1G and 2M huge pages have different allocation and initialization logic,
which leads to subtle differences in parallelization. Therefore, it is
appropriate to split hugetlb_hstate_alloc_pages into gigantic and
non-gigantic.

This patch has no functional changes.

Signed-off-by: Gang Li <gang.li@linux.dev>
---
 mm/hugetlb.c | 86 +++++++++++++++++++++++++++-------------------------
 1 file changed, 45 insertions(+), 41 deletions(-)

Comments

Tim Chen Jan. 10, 2024, 11:12 p.m. UTC | #1 
On Tue, 2024-01-02 at 21:12 +0800, Gang Li wrote:
> 1G and 2M huge pages have different allocation and initialization logic,
> which leads to subtle differences in parallelization. Therefore, it is
> appropriate to split hugetlb_hstate_alloc_pages into gigantic and
> non-gigantic.
> 
> This patch has no functional changes.
> 
> Signed-off-by: Gang Li <gang.li@linux.dev>
> ---
>  mm/hugetlb.c | 86 +++++++++++++++++++++++++++-------------------------
>  1 file changed, 45 insertions(+), 41 deletions(-)
> 
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 2606135ec55e6..92448e747991d 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -3509,6 +3509,47 @@ static void __init hugetlb_hstate_alloc_pages_report(unsigned long allocated, st
>  	}
>  }
>  
> +static unsigned long __init hugetlb_hstate_alloc_pages_gigantic(struct hstate *h)
> +{
> +	unsigned long i;
> +
> +	for (i = 0; i < h->max_huge_pages; ++i) {
> +		/*
> +		 * gigantic pages not added to list as they are not
> +		 * added to pools now.
> +		 */

This comment unnecessary as now we don't have mix gigantic and non-gigantic code,
which uses foilio list.  And folio_list is not in this routine.

Can be removed.

Otherwise Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>

> +		if (!alloc_bootmem_huge_page(h, NUMA_NO_NODE))
> +			break;
> +		cond_resched();
> +	}
> +
> +	return i;
> +}
> +
> +static unsigned long __init hugetlb_hstate_alloc_pages_non_gigantic(struct hstate *h)
> +{
> +	unsigned long i;
> +	struct folio *folio;
> +	LIST_HEAD(folio_list);
> +	nodemask_t node_alloc_noretry;
> +
> +	/* Bit mask controlling how hard we retry per-node allocations.*/
> +	nodes_clear(node_alloc_noretry);
> +
> +	for (i = 0; i < h->max_huge_pages; ++i) {
> +		folio = alloc_pool_huge_folio(h, &node_states[N_MEMORY],
> +						&node_alloc_noretry);
> +		if (!folio)
> +			break;
> +		list_add(&folio->lru, &folio_list);
> +		cond_resched();
> +	}
> +
> +	prep_and_add_allocated_folios(h, &folio_list);
> +
> +	return i;
> +}
> +
>  /*
>   * NOTE: this routine is called in different contexts for gigantic and
>   * non-gigantic pages.
> @@ -3522,10 +3563,7 @@ static void __init hugetlb_hstate_alloc_pages_report(unsigned long allocated, st
>   */
>  static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
>  {
> -	unsigned long i;
> -	struct folio *folio;
> -	LIST_HEAD(folio_list);
> -	nodemask_t *node_alloc_noretry;
> +	unsigned long allocated;
>  
>  	/* skip gigantic hugepages allocation if hugetlb_cma enabled */
>  	if (hstate_is_gigantic(h) && hugetlb_cma_size) {
> @@ -3539,46 +3577,12 @@ static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
>  
>  	/* below will do all node balanced alloc */
>  	if (!hstate_is_gigantic(h)) {
> -		/*
> -		 * Bit mask controlling how hard we retry per-node allocations.
> -		 * Ignore errors as lower level routines can deal with
> -		 * node_alloc_noretry == NULL.  If this kmalloc fails at boot
> -		 * time, we are likely in bigger trouble.
> -		 */
> -		node_alloc_noretry = kmalloc(sizeof(*node_alloc_noretry),
> -						GFP_KERNEL);
> +		allocated = hugetlb_hstate_alloc_pages_non_gigantic(h);
>  	} else {
> -		/* allocations done at boot time */
> -		node_alloc_noretry = NULL;
> -	}
> -
> -	/* bit mask controlling how hard we retry per-node allocations */
> -	if (node_alloc_noretry)
> -		nodes_clear(*node_alloc_noretry);
> -
> -	for (i = 0; i < h->max_huge_pages; ++i) {
> -		if (hstate_is_gigantic(h)) {
> -			/*
> -			 * gigantic pages not added to list as they are not
> -			 * added to pools now.
> -			 */
> -			if (!alloc_bootmem_huge_page(h, NUMA_NO_NODE))
> -				break;
> -		} else {
> -			folio = alloc_pool_huge_folio(h, &node_states[N_MEMORY],
> -							node_alloc_noretry);
> -			if (!folio)
> -				break;
> -			list_add(&folio->lru, &folio_list);
> -		}
> -		cond_resched();
> +		allocated = hugetlb_hstate_alloc_pages_gigantic(h);
>  	}
>  
> -	/* list will be empty if hstate_is_gigantic */
> -	prep_and_add_allocated_folios(h, &folio_list);
> -
> -	hugetlb_hstate_alloc_pages_report(i, h);
> -	kfree(node_alloc_noretry);
> +	hugetlb_hstate_alloc_pages_report(allocated, h);
>  }
>  
>  static void __init hugetlb_init_hstates(void)
Gang Li Jan. 11, 2024, 3:44 a.m. UTC | #2 
On 2024/1/11 07:12, Tim Chen wrote:
> On Tue, 2024-01-02 at 21:12 +0800, Gang Li wrote:
>> +static unsigned long __init hugetlb_hstate_alloc_pages_gigantic(struct hstate *h)
>> +{
>> +	unsigned long i;
>> +
>> +	for (i = 0; i < h->max_huge_pages; ++i) {
>> +		/*
>> +		 * gigantic pages not added to list as they are not
>> +		 * added to pools now.
>> +		 */
> 
> This comment unnecessary as now we don't have mix gigantic and non-gigantic code,
> which uses foilio list.  And folio_list is not in this routine.
> 
> Can be removed.
> 
> Otherwise Reviewed-by: Tim Chen <tim.c.chen@linux.intel.com>
> 

Thanks!
Muchun Song Jan. 16, 2024, 7:02 a.m. UTC | #3 
On 2024/1/2 21:12, Gang Li wrote:
> 1G and 2M huge pages have different allocation and initialization logic,
> which leads to subtle differences in parallelization. Therefore, it is
> appropriate to split hugetlb_hstate_alloc_pages into gigantic and
> non-gigantic.
>
> This patch has no functional changes.
>
> Signed-off-by: Gang Li <gang.li@linux.dev>
> ---
>   mm/hugetlb.c | 86 +++++++++++++++++++++++++++-------------------------
>   1 file changed, 45 insertions(+), 41 deletions(-)
>
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 2606135ec55e6..92448e747991d 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -3509,6 +3509,47 @@ static void __init hugetlb_hstate_alloc_pages_report(unsigned long allocated, st
>   	}
>   }
>   
> +static unsigned long __init hugetlb_hstate_alloc_pages_gigantic(struct hstate *h)

The name is so long, how about hugetlb_gigantic_pages_alloc_boot?

> +{
> +	unsigned long i;
> +
> +	for (i = 0; i < h->max_huge_pages; ++i) {
> +		/*
> +		 * gigantic pages not added to list as they are not
> +		 * added to pools now.
> +		 */
> +		if (!alloc_bootmem_huge_page(h, NUMA_NO_NODE))
> +			break;
> +		cond_resched();
> +	}
> +
> +	return i;
> +}
> +
> +static unsigned long __init hugetlb_hstate_alloc_pages_non_gigantic(struct hstate *h)

hugetlb_pages_alloc_boot?

> +{
> +	unsigned long i;
> +	struct folio *folio;
> +	LIST_HEAD(folio_list);
> +	nodemask_t node_alloc_noretry;
> +
> +	/* Bit mask controlling how hard we retry per-node allocations.*/
> +	nodes_clear(node_alloc_noretry);
> +
> +	for (i = 0; i < h->max_huge_pages; ++i) {
> +		folio = alloc_pool_huge_folio(h, &node_states[N_MEMORY],
> +						&node_alloc_noretry);
> +		if (!folio)
> +			break;
> +		list_add(&folio->lru, &folio_list);
> +		cond_resched();
> +	}
> +
> +	prep_and_add_allocated_folios(h, &folio_list);
> +
> +	return i;
> +}
> +
>   /*
>    * NOTE: this routine is called in different contexts for gigantic and
>    * non-gigantic pages.
> @@ -3522,10 +3563,7 @@ static void __init hugetlb_hstate_alloc_pages_report(unsigned long allocated, st
>    */
>   static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
>   {
> -	unsigned long i;
> -	struct folio *folio;
> -	LIST_HEAD(folio_list);
> -	nodemask_t *node_alloc_noretry;
> +	unsigned long allocated;
>   
>   	/* skip gigantic hugepages allocation if hugetlb_cma enabled */
>   	if (hstate_is_gigantic(h) && hugetlb_cma_size) {
> @@ -3539,46 +3577,12 @@ static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
>   
>   	/* below will do all node balanced alloc */
>   	if (!hstate_is_gigantic(h)) {

It it unnecessary to reverse the condition. A little sime like following:

if (hstate_is_gigantic(h))
     /* gigantic pages */
else
     /* normal pages */

> -		/*
> -		 * Bit mask controlling how hard we retry per-node allocations.
> -		 * Ignore errors as lower level routines can deal with
> -		 * node_alloc_noretry == NULL.  If this kmalloc fails at boot
> -		 * time, we are likely in bigger trouble.
> -		 */
> -		node_alloc_noretry = kmalloc(sizeof(*node_alloc_noretry),
> -						GFP_KERNEL);
> +		allocated = hugetlb_hstate_alloc_pages_non_gigantic(h);
>   	} else {
> -		/* allocations done at boot time */
> -		node_alloc_noretry = NULL;
> -	}
> -
> -	/* bit mask controlling how hard we retry per-node allocations */
> -	if (node_alloc_noretry)
> -		nodes_clear(*node_alloc_noretry);
> -
> -	for (i = 0; i < h->max_huge_pages; ++i) {
> -		if (hstate_is_gigantic(h)) {
> -			/*
> -			 * gigantic pages not added to list as they are not
> -			 * added to pools now.
> -			 */
> -			if (!alloc_bootmem_huge_page(h, NUMA_NO_NODE))
> -				break;
> -		} else {
> -			folio = alloc_pool_huge_folio(h, &node_states[N_MEMORY],
> -							node_alloc_noretry);
> -			if (!folio)
> -				break;
> -			list_add(&folio->lru, &folio_list);
> -		}
> -		cond_resched();
> +		allocated = hugetlb_hstate_alloc_pages_gigantic(h);
>   	}
>   
> -	/* list will be empty if hstate_is_gigantic */
> -	prep_and_add_allocated_folios(h, &folio_list);
> -
> -	hugetlb_hstate_alloc_pages_report(i, h);
> -	kfree(node_alloc_noretry);
> +	hugetlb_hstate_alloc_pages_report(allocated, h);
>   }
>   
>   static void __init hugetlb_init_hstates(void)
Gang Li Jan. 16, 2024, 8:09 a.m. UTC | #4 
On 2024/1/16 15:02, Muchun Song wrote:
> On 2024/1/2 21:12, Gang Li wrote:
>> hugetlb_hstate_alloc_pages_gigantic(struct hstate *h)
> 
> The name is so long, how about hugetlb_gigantic_pages_alloc_boot?
> 
>> hugetlb_hstate_alloc_pages_non_gigantic(struct hstate *h)
> 
> hugetlb_pages_alloc_boot?
LGTM.

> It it unnecessary to reverse the condition. A little sime like following:
> 
> if (hstate_is_gigantic(h))
>     /* gigantic pages */
> else
>     /* normal pages */

Will take it in next version.
diff mbox series

Patch

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 2606135ec55e6..92448e747991d 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -3509,6 +3509,47 @@  static void __init hugetlb_hstate_alloc_pages_report(unsigned long allocated, st
 	}
 }
 
+static unsigned long __init hugetlb_hstate_alloc_pages_gigantic(struct hstate *h)
+{
+	unsigned long i;
+
+	for (i = 0; i < h->max_huge_pages; ++i) {
+		/*
+		 * gigantic pages not added to list as they are not
+		 * added to pools now.
+		 */
+		if (!alloc_bootmem_huge_page(h, NUMA_NO_NODE))
+			break;
+		cond_resched();
+	}
+
+	return i;
+}
+
+static unsigned long __init hugetlb_hstate_alloc_pages_non_gigantic(struct hstate *h)
+{
+	unsigned long i;
+	struct folio *folio;
+	LIST_HEAD(folio_list);
+	nodemask_t node_alloc_noretry;
+
+	/* Bit mask controlling how hard we retry per-node allocations.*/
+	nodes_clear(node_alloc_noretry);
+
+	for (i = 0; i < h->max_huge_pages; ++i) {
+		folio = alloc_pool_huge_folio(h, &node_states[N_MEMORY],
+						&node_alloc_noretry);
+		if (!folio)
+			break;
+		list_add(&folio->lru, &folio_list);
+		cond_resched();
+	}
+
+	prep_and_add_allocated_folios(h, &folio_list);
+
+	return i;
+}
+
 /*
  * NOTE: this routine is called in different contexts for gigantic and
  * non-gigantic pages.
@@ -3522,10 +3563,7 @@  static void __init hugetlb_hstate_alloc_pages_report(unsigned long allocated, st
  */
 static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
 {
-	unsigned long i;
-	struct folio *folio;
-	LIST_HEAD(folio_list);
-	nodemask_t *node_alloc_noretry;
+	unsigned long allocated;
 
 	/* skip gigantic hugepages allocation if hugetlb_cma enabled */
 	if (hstate_is_gigantic(h) && hugetlb_cma_size) {
@@ -3539,46 +3577,12 @@  static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
 
 	/* below will do all node balanced alloc */
 	if (!hstate_is_gigantic(h)) {
-		/*
-		 * Bit mask controlling how hard we retry per-node allocations.
-		 * Ignore errors as lower level routines can deal with
-		 * node_alloc_noretry == NULL.  If this kmalloc fails at boot
-		 * time, we are likely in bigger trouble.
-		 */
-		node_alloc_noretry = kmalloc(sizeof(*node_alloc_noretry),
-						GFP_KERNEL);
+		allocated = hugetlb_hstate_alloc_pages_non_gigantic(h);
 	} else {
-		/* allocations done at boot time */
-		node_alloc_noretry = NULL;
-	}
-
-	/* bit mask controlling how hard we retry per-node allocations */
-	if (node_alloc_noretry)
-		nodes_clear(*node_alloc_noretry);
-
-	for (i = 0; i < h->max_huge_pages; ++i) {
-		if (hstate_is_gigantic(h)) {
-			/*
-			 * gigantic pages not added to list as they are not
-			 * added to pools now.
-			 */
-			if (!alloc_bootmem_huge_page(h, NUMA_NO_NODE))
-				break;
-		} else {
-			folio = alloc_pool_huge_folio(h, &node_states[N_MEMORY],
-							node_alloc_noretry);
-			if (!folio)
-				break;
-			list_add(&folio->lru, &folio_list);
-		}
-		cond_resched();
+		allocated = hugetlb_hstate_alloc_pages_gigantic(h);
 	}
 
-	/* list will be empty if hstate_is_gigantic */
-	prep_and_add_allocated_folios(h, &folio_list);
-
-	hugetlb_hstate_alloc_pages_report(i, h);
-	kfree(node_alloc_noretry);
+	hugetlb_hstate_alloc_pages_report(allocated, h);
 }
 
 static void __init hugetlb_init_hstates(void)