[PATCH 0/6] A few cleanup patches for hugetlb_cgroup

[PATCH 0/6] A few cleanup patches for hugetlb_cgroup

[Miaohe Lin]

Hi everyone,
This series contains a few cleaup patches to remove unneeded header file,
use helper macro, remove unneeded return value and so on. More details
can be found in the respective changelogs.
Thanks!

Miaohe Lin (6):
  hugetlb_cgroup: remove unneeded nr_pages > 0 check
  hugetlb_cgroup: remove unneeded header file
  hugetlb_cgroup: hugetlbfs: use helper macro SZ_1{K,M,G}
  hugetlb_cgroup: remove unneeded return value
  hugetlb_cgroup: use helper macro NUMA_NO_NODE
  hugetlb_cgroup: use helper for_each_hstate and hstate_index

 include/linux/hugetlb_cgroup.h | 19 ++++++++-----------
 mm/hugetlb_cgroup.c            | 28 ++++++++++++----------------
 2 files changed, 20 insertions(+), 27 deletions(-)

-- 
2.23.0

[PATCH 1/6] hugetlb_cgroup: remove unneeded nr_pages > 0 check

[Miaohe Lin]

When code reaches here, nr_pages must be > 0. Remove unneeded nr_pages > 0
check to simplify the code.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/hugetlb_cgroup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
index c86691c431fd..d16eb00c947d 100644
--- a/mm/hugetlb_cgroup.c
+++ b/mm/hugetlb_cgroup.c
@@ -442,7 +442,7 @@ void hugetlb_cgroup_uncharge_file_region(struct resv_map *resv,
 	if (hugetlb_cgroup_disabled() || !resv || !rg || !nr_pages)
 		return;
 
-	if (rg->reservation_counter && resv->pages_per_hpage && nr_pages > 0 &&
+	if (rg->reservation_counter && resv->pages_per_hpage &&
 	    !resv->reservation_counter) {
 		page_counter_uncharge(rg->reservation_counter,
 				      nr_pages * resv->pages_per_hpage);
-- 
2.23.0

Re: [PATCH 1/6] hugetlb_cgroup: remove unneeded nr_pages > 0 check

[Mina Almasry]

On Thu, Jul 28, 2022 at 5:20 AM Miaohe Lin <linmiaohe@huawei.com> wrote:
>
> When code reaches here, nr_pages must be > 0. Remove unneeded nr_pages > 0
> check to simplify the code.
>
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>

Thanks Miaohe,

Reviewed-by: Mina Almasry <almasrymina@google.com>

> ---
>  mm/hugetlb_cgroup.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
> index c86691c431fd..d16eb00c947d 100644
> --- a/mm/hugetlb_cgroup.c
> +++ b/mm/hugetlb_cgroup.c
> @@ -442,7 +442,7 @@ void hugetlb_cgroup_uncharge_file_region(struct resv_map *resv,
>         if (hugetlb_cgroup_disabled() || !resv || !rg || !nr_pages)
>                 return;
>
> -       if (rg->reservation_counter && resv->pages_per_hpage && nr_pages > 0 &&
> +       if (rg->reservation_counter && resv->pages_per_hpage &&
>             !resv->reservation_counter) {
>                 page_counter_uncharge(rg->reservation_counter,
>                                       nr_pages * resv->pages_per_hpage);
> --
> 2.23.0
>

[PATCH 2/6] hugetlb_cgroup: remove unneeded header file

[Miaohe Lin]

The header file slab.h is unneeded now. Remove it.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/hugetlb_cgroup.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
index d16eb00c947d..99e9a367e1e5 100644
--- a/mm/hugetlb_cgroup.c
+++ b/mm/hugetlb_cgroup.c
@@ -19,7 +19,6 @@
 
 #include <linux/cgroup.h>
 #include <linux/page_counter.h>
-#include <linux/slab.h>
 #include <linux/hugetlb.h>
 #include <linux/hugetlb_cgroup.h>
 
-- 
2.23.0

Re: [PATCH 2/6] hugetlb_cgroup: remove unneeded header file

[Mina Almasry]

On Thu, Jul 28, 2022 at 5:20 AM Miaohe Lin <linmiaohe@huawei.com> wrote:
>
> The header file slab.h is unneeded now. Remove it.
>
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>

Acked-by: Mina Almasry <almasrymina@google.com>

> ---
>  mm/hugetlb_cgroup.c | 1 -
>  1 file changed, 1 deletion(-)
>
> diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
> index d16eb00c947d..99e9a367e1e5 100644
> --- a/mm/hugetlb_cgroup.c
> +++ b/mm/hugetlb_cgroup.c
> @@ -19,7 +19,6 @@
>
>  #include <linux/cgroup.h>
>  #include <linux/page_counter.h>
> -#include <linux/slab.h>
>  #include <linux/hugetlb.h>
>  #include <linux/hugetlb_cgroup.h>
>
> --
> 2.23.0
>

Re: [PATCH 2/6] hugetlb_cgroup: remove unneeded header file

[Mina Almasry]

On Thu, Jul 28, 2022 at 10:45 AM Mina Almasry <almasrymina@google.com> wrote:
>
> On Thu, Jul 28, 2022 at 5:20 AM Miaohe Lin <linmiaohe@huawei.com> wrote:
> >
> > The header file slab.h is unneeded now. Remove it.
> >
> > Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
>
> Acked-by: Mina Almasry <almasrymina@google.com>
>

Actually, sorry, but this file calls kzalloc_node() which comes from
linux/slab.h, no? Can we keep the include then? slab.h is probably
included by other include files but for clarity it's better that
hugetlb_cgroup.c to include slab.h if it is calling functions declared
there.

> > ---
> >  mm/hugetlb_cgroup.c | 1 -
> >  1 file changed, 1 deletion(-)
> >
> > diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
> > index d16eb00c947d..99e9a367e1e5 100644
> > --- a/mm/hugetlb_cgroup.c
> > +++ b/mm/hugetlb_cgroup.c
> > @@ -19,7 +19,6 @@
> >
> >  #include <linux/cgroup.h>
> >  #include <linux/page_counter.h>
> > -#include <linux/slab.h>
> >  #include <linux/hugetlb.h>
> >  #include <linux/hugetlb_cgroup.h>
> >
> > --
> > 2.23.0
> >

Re: [PATCH 2/6] hugetlb_cgroup: remove unneeded header file

[Miaohe Lin]

On 2022/7/29 2:02, Mina Almasry wrote:
> On Thu, Jul 28, 2022 at 10:45 AM Mina Almasry <almasrymina@google.com> wrote:
>>
>> On Thu, Jul 28, 2022 at 5:20 AM Miaohe Lin <linmiaohe@huawei.com> wrote:
>>>
>>> The header file slab.h is unneeded now. Remove it.
>>>
>>> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
>>
>> Acked-by: Mina Almasry <almasrymina@google.com>
>>
> 
> Actually, sorry, but this file calls kzalloc_node() which comes from
> linux/slab.h, no? Can we keep the include then? slab.h is probably
> included by other include files but for clarity it's better that
> hugetlb_cgroup.c to include slab.h if it is calling functions declared
> there.

I have browsed the code to verify it but I still missed kzalloc_node in this file.
Will drop this patch. Many thanks for your review.

> 
>>> ---
>>>  mm/hugetlb_cgroup.c | 1 -
>>>  1 file changed, 1 deletion(-)
>>>
>>> diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
>>> index d16eb00c947d..99e9a367e1e5 100644
>>> --- a/mm/hugetlb_cgroup.c
>>> +++ b/mm/hugetlb_cgroup.c
>>> @@ -19,7 +19,6 @@
>>>
>>>  #include <linux/cgroup.h>
>>>  #include <linux/page_counter.h>
>>> -#include <linux/slab.h>
>>>  #include <linux/hugetlb.h>
>>>  #include <linux/hugetlb_cgroup.h>
>>>
>>> --
>>> 2.23.0
>>>
> .
> 

[PATCH 3/6] hugetlb_cgroup: hugetlbfs: use helper macro SZ_1{K,M,G}

[Miaohe Lin]

Use helper macro SZ_1K, SZ_1M and SZ_1G to do the size conversion. Minor
readability improvement.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/hugetlb_cgroup.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
index 99e9a367e1e5..7e0bca52c40f 100644
--- a/mm/hugetlb_cgroup.c
+++ b/mm/hugetlb_cgroup.c
@@ -674,12 +674,12 @@ static ssize_t hugetlb_cgroup_reset(struct kernfs_open_file *of,
 
 static char *mem_fmt(char *buf, int size, unsigned long hsize)
 {
-	if (hsize >= (1UL << 30))
-		snprintf(buf, size, "%luGB", hsize >> 30);
-	else if (hsize >= (1UL << 20))
-		snprintf(buf, size, "%luMB", hsize >> 20);
+	if (hsize >= SZ_1G)
+		snprintf(buf, size, "%luGB", hsize / SZ_1G);
+	else if (hsize >= SZ_1M)
+		snprintf(buf, size, "%luMB", hsize / SZ_1M);
 	else
-		snprintf(buf, size, "%luKB", hsize >> 10);
+		snprintf(buf, size, "%luKB", hsize / SZ_1K);
 	return buf;
 }
 
-- 
2.23.0

Re: [PATCH 3/6] hugetlb_cgroup: hugetlbfs: use helper macro SZ_1{K,M,G}

[Mina Almasry]

On Thu, Jul 28, 2022 at 5:20 AM Miaohe Lin <linmiaohe@huawei.com> wrote:
>
> Use helper macro SZ_1K, SZ_1M and SZ_1G to do the size conversion. Minor
> readability improvement.
>
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>

Thanks Miaohe,

Reviewed-by: Mina Almasry <almasrymina@google.com>

> ---
>  mm/hugetlb_cgroup.c | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
> index 99e9a367e1e5..7e0bca52c40f 100644
> --- a/mm/hugetlb_cgroup.c
> +++ b/mm/hugetlb_cgroup.c
> @@ -674,12 +674,12 @@ static ssize_t hugetlb_cgroup_reset(struct kernfs_open_file *of,
>
>  static char *mem_fmt(char *buf, int size, unsigned long hsize)
>  {
> -       if (hsize >= (1UL << 30))
> -               snprintf(buf, size, "%luGB", hsize >> 30);
> -       else if (hsize >= (1UL << 20))
> -               snprintf(buf, size, "%luMB", hsize >> 20);
> +       if (hsize >= SZ_1G)
> +               snprintf(buf, size, "%luGB", hsize / SZ_1G);
> +       else if (hsize >= SZ_1M)
> +               snprintf(buf, size, "%luMB", hsize / SZ_1M);
>         else
> -               snprintf(buf, size, "%luKB", hsize >> 10);
> +               snprintf(buf, size, "%luKB", hsize / SZ_1K);
>         return buf;
>  }
>
> --
> 2.23.0
>

[PATCH 4/6] hugetlb_cgroup: remove unneeded return value

[Miaohe Lin]

The return value of set_hugetlb_cgroup and set_hugetlb_cgroup_rsvd are
always ignored. Remove them to clean up the code.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 include/linux/hugetlb_cgroup.h | 19 ++++++++-----------
 1 file changed, 8 insertions(+), 11 deletions(-)

diff --git a/include/linux/hugetlb_cgroup.h b/include/linux/hugetlb_cgroup.h
index 379344828e78..630cd255d0cf 100644
--- a/include/linux/hugetlb_cgroup.h
+++ b/include/linux/hugetlb_cgroup.h
@@ -90,32 +90,31 @@ hugetlb_cgroup_from_page_rsvd(struct page *page)
 	return __hugetlb_cgroup_from_page(page, true);
 }
 
-static inline int __set_hugetlb_cgroup(struct page *page,
+static inline void __set_hugetlb_cgroup(struct page *page,
 				       struct hugetlb_cgroup *h_cg, bool rsvd)
 {
 	VM_BUG_ON_PAGE(!PageHuge(page), page);
 
 	if (compound_order(page) < HUGETLB_CGROUP_MIN_ORDER)
-		return -1;
+		return;
 	if (rsvd)
 		set_page_private(page + SUBPAGE_INDEX_CGROUP_RSVD,
 				 (unsigned long)h_cg);
 	else
 		set_page_private(page + SUBPAGE_INDEX_CGROUP,
 				 (unsigned long)h_cg);
-	return 0;
 }
 
-static inline int set_hugetlb_cgroup(struct page *page,
+static inline void set_hugetlb_cgroup(struct page *page,
 				     struct hugetlb_cgroup *h_cg)
 {
-	return __set_hugetlb_cgroup(page, h_cg, false);
+	__set_hugetlb_cgroup(page, h_cg, false);
 }
 
-static inline int set_hugetlb_cgroup_rsvd(struct page *page,
+static inline void set_hugetlb_cgroup_rsvd(struct page *page,
 					  struct hugetlb_cgroup *h_cg)
 {
-	return __set_hugetlb_cgroup(page, h_cg, true);
+	__set_hugetlb_cgroup(page, h_cg, true);
 }
 
 static inline bool hugetlb_cgroup_disabled(void)
@@ -199,16 +198,14 @@ hugetlb_cgroup_from_page_rsvd(struct page *page)
 	return NULL;
 }
 
-static inline int set_hugetlb_cgroup(struct page *page,
+static inline void set_hugetlb_cgroup(struct page *page,
 				     struct hugetlb_cgroup *h_cg)
 {
-	return 0;
 }
 
-static inline int set_hugetlb_cgroup_rsvd(struct page *page,
+static inline void set_hugetlb_cgroup_rsvd(struct page *page,
 					  struct hugetlb_cgroup *h_cg)
 {
-	return 0;
 }
 
 static inline bool hugetlb_cgroup_disabled(void)
-- 
2.23.0

Re: [PATCH 4/6] hugetlb_cgroup: remove unneeded return value

[Mike Kravetz]

On 07/28/22 20:19, Miaohe Lin wrote:
> The return value of set_hugetlb_cgroup and set_hugetlb_cgroup_rsvd are
> always ignored. Remove them to clean up the code.

Agree.  Thanks!

> 
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
> ---
>  include/linux/hugetlb_cgroup.h | 19 ++++++++-----------
>  1 file changed, 8 insertions(+), 11 deletions(-)

Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
-- 
Mike Kravetz

[PATCH 5/6] hugetlb_cgroup: use helper macro NUMA_NO_NODE

[Miaohe Lin]

It's better to use NUMA_NO_NODE instead of magic number -1. Minor
readability improvement.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/hugetlb_cgroup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
index 7e0bca52c40f..13ec091385af 100644
--- a/mm/hugetlb_cgroup.c
+++ b/mm/hugetlb_cgroup.c
@@ -153,9 +153,9 @@ hugetlb_cgroup_css_alloc(struct cgroup_subsys_state *parent_css)
 	 * function.
 	 */
 	for_each_node(node) {
-		/* Set node_to_alloc to -1 for offline nodes. */
+		/* Set node_to_alloc to NUMA_NO_NODE for offline nodes. */
 		int node_to_alloc =
-			node_state(node, N_NORMAL_MEMORY) ? node : -1;
+			node_state(node, N_NORMAL_MEMORY) ? node : NUMA_NO_NODE;
 		h_cgroup->nodeinfo[node] =
 			kzalloc_node(sizeof(struct hugetlb_cgroup_per_node),
 				     GFP_KERNEL, node_to_alloc);
-- 
2.23.0

Re: [PATCH 5/6] hugetlb_cgroup: use helper macro NUMA_NO_NODE

[Mina Almasry]

On Thu, Jul 28, 2022 at 5:20 AM Miaohe Lin <linmiaohe@huawei.com> wrote:
>
> It's better to use NUMA_NO_NODE instead of magic number -1. Minor
> readability improvement.
>
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>

Reviewed-by: Mina Almasry <almasrymina@google.com>

> ---
>  mm/hugetlb_cgroup.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
> index 7e0bca52c40f..13ec091385af 100644
> --- a/mm/hugetlb_cgroup.c
> +++ b/mm/hugetlb_cgroup.c
> @@ -153,9 +153,9 @@ hugetlb_cgroup_css_alloc(struct cgroup_subsys_state *parent_css)
>          * function.
>          */
>         for_each_node(node) {
> -               /* Set node_to_alloc to -1 for offline nodes. */
> +               /* Set node_to_alloc to NUMA_NO_NODE for offline nodes. */
>                 int node_to_alloc =
> -                       node_state(node, N_NORMAL_MEMORY) ? node : -1;
> +                       node_state(node, N_NORMAL_MEMORY) ? node : NUMA_NO_NODE;
>                 h_cgroup->nodeinfo[node] =
>                         kzalloc_node(sizeof(struct hugetlb_cgroup_per_node),
>                                      GFP_KERNEL, node_to_alloc);
> --
> 2.23.0
>

[PATCH 6/6] hugetlb_cgroup: use helper for_each_hstate and hstate_index

[Miaohe Lin]

Use helper for_each_hstate and hstate_index to iterate the hstate and get
the hstate index. Minor readability improvement.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/hugetlb_cgroup.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
index 13ec091385af..8f20082df4c8 100644
--- a/mm/hugetlb_cgroup.c
+++ b/mm/hugetlb_cgroup.c
@@ -74,11 +74,11 @@ parent_hugetlb_cgroup(struct hugetlb_cgroup *h_cg)
 
 static inline bool hugetlb_cgroup_have_usage(struct hugetlb_cgroup *h_cg)
 {
-	int idx;
+	struct hstate *h;
 
-	for (idx = 0; idx < hugetlb_max_hstate; idx++) {
+	for_each_hstate(h) {
 		if (page_counter_read(
-				hugetlb_cgroup_counter_from_cgroup(h_cg, idx)))
+		    hugetlb_cgroup_counter_from_cgroup(h_cg, hstate_index(h))))
 			return true;
 	}
 	return false;
@@ -224,17 +224,14 @@ static void hugetlb_cgroup_css_offline(struct cgroup_subsys_state *css)
 	struct hugetlb_cgroup *h_cg = hugetlb_cgroup_from_css(css);
 	struct hstate *h;
 	struct page *page;
-	int idx;
 
 	do {
-		idx = 0;
 		for_each_hstate(h) {
 			spin_lock_irq(&hugetlb_lock);
 			list_for_each_entry(page, &h->hugepage_activelist, lru)
-				hugetlb_cgroup_move_parent(idx, h_cg, page);
+				hugetlb_cgroup_move_parent(hstate_index(h), h_cg, page);
 
 			spin_unlock_irq(&hugetlb_lock);
-			idx++;
 		}
 		cond_resched();
 	} while (hugetlb_cgroup_have_usage(h_cg));
-- 
2.23.0

Re: [PATCH 6/6] hugetlb_cgroup: use helper for_each_hstate and hstate_index

[Mina Almasry]

On Thu, Jul 28, 2022 at 5:20 AM Miaohe Lin <linmiaohe@huawei.com> wrote:
>
> Use helper for_each_hstate and hstate_index to iterate the hstate and get
> the hstate index. Minor readability improvement.
>
> Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>

Reviewed-by: Mina Almasry <almasrymina@google.com>

> ---
>  mm/hugetlb_cgroup.c | 11 ++++-------
>  1 file changed, 4 insertions(+), 7 deletions(-)
>
> diff --git a/mm/hugetlb_cgroup.c b/mm/hugetlb_cgroup.c
> index 13ec091385af..8f20082df4c8 100644
> --- a/mm/hugetlb_cgroup.c
> +++ b/mm/hugetlb_cgroup.c
> @@ -74,11 +74,11 @@ parent_hugetlb_cgroup(struct hugetlb_cgroup *h_cg)
>
>  static inline bool hugetlb_cgroup_have_usage(struct hugetlb_cgroup *h_cg)
>  {
> -       int idx;
> +       struct hstate *h;
>
> -       for (idx = 0; idx < hugetlb_max_hstate; idx++) {
> +       for_each_hstate(h) {
>                 if (page_counter_read(
> -                               hugetlb_cgroup_counter_from_cgroup(h_cg, idx)))
> +                   hugetlb_cgroup_counter_from_cgroup(h_cg, hstate_index(h))))
>                         return true;
>         }
>         return false;
> @@ -224,17 +224,14 @@ static void hugetlb_cgroup_css_offline(struct cgroup_subsys_state *css)
>         struct hugetlb_cgroup *h_cg = hugetlb_cgroup_from_css(css);
>         struct hstate *h;
>         struct page *page;
> -       int idx;
>
>         do {
> -               idx = 0;
>                 for_each_hstate(h) {
>                         spin_lock_irq(&hugetlb_lock);
>                         list_for_each_entry(page, &h->hugepage_activelist, lru)
> -                               hugetlb_cgroup_move_parent(idx, h_cg, page);
> +                               hugetlb_cgroup_move_parent(hstate_index(h), h_cg, page);
>
>                         spin_unlock_irq(&hugetlb_lock);
> -                       idx++;
>                 }
>                 cond_resched();
>         } while (hugetlb_cgroup_have_usage(h_cg));
> --
> 2.23.0
>