fs/proc/page.c | 166 +++++++++++++++++-------------------------------- 1 file changed, 58 insertions(+), 108 deletions(-)
From: Liu Ye <liuye@kylinos.cn>
The function kpageflags_read and kpagecgroup_read is quite similar
to kpagecount_read. Consider refactoring common code into a helper
function to reduce code duplication.
Signed-off-by: Liu Ye <liuye@kylinos.cn>
---
V2 : Use an enumeration to indicate the operation to be performed
to avoid passing functions.
---
---
fs/proc/page.c | 166 +++++++++++++++++--------------------------------
1 file changed, 58 insertions(+), 108 deletions(-)
diff --git a/fs/proc/page.c b/fs/proc/page.c
index a55f5acefa97..66f454330a87 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -22,6 +22,14 @@
#define KPMMASK (KPMSIZE - 1)
#define KPMBITS (KPMSIZE * BITS_PER_BYTE)
+enum kpage_operation {
+ KPAGE_FLAGS,
+ KPAGE_COUNT,
+#ifdef CONFIG_MEMCG
+ KPAGE_CGROUP,
+#endif
+};
+
static inline unsigned long get_max_dump_pfn(void)
{
#ifdef CONFIG_SPARSEMEM
@@ -37,19 +45,17 @@ static inline unsigned long get_max_dump_pfn(void)
#endif
}
-/* /proc/kpagecount - an array exposing page mapcounts
- *
- * Each entry is a u64 representing the corresponding
- * physical page mapcount.
- */
-static ssize_t kpagecount_read(struct file *file, char __user *buf,
- size_t count, loff_t *ppos)
+static ssize_t kpage_read(struct file *file, char __user *buf,
+ size_t count, loff_t *ppos,
+ enum kpage_operation op)
{
const unsigned long max_dump_pfn = get_max_dump_pfn();
u64 __user *out = (u64 __user *)buf;
+ struct page *ppage;
unsigned long src = *ppos;
unsigned long pfn;
ssize_t ret = 0;
+ u64 info;
pfn = src / KPMSIZE;
if (src & KPMMASK || count & KPMMASK)
@@ -59,19 +65,29 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
while (count > 0) {
- struct page *page;
- u64 mapcount = 0;
-
- /*
- * TODO: ZONE_DEVICE support requires to identify
- * memmaps that were actually initialized.
- */
- page = pfn_to_online_page(pfn);
- if (page)
- mapcount = folio_precise_page_mapcount(page_folio(page),
- page);
-
- if (put_user(mapcount, out)) {
+ ppage = pfn_to_online_page(pfn);
+
+ if (ppage) {
+ switch (op) {
+ case KPAGE_FLAGS:
+ info = stable_page_flags(ppage);
+ break;
+ case KPAGE_COUNT:
+ info = folio_precise_page_mapcount(page_folio(ppage), ppage);
+ break;
+#ifdef CONFIG_MEMCG
+ case KPAGE_CGROUP:
+ info = page_cgroup_ino(ppage);
+ break;
+#endif
+ default:
+ info = 0;
+ break;
+ }
+ } else
+ info = 0;
+
+ if (put_user(info, out)) {
ret = -EFAULT;
break;
}
@@ -89,17 +105,23 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
return ret;
}
+/* /proc/kpagecount - an array exposing page mapcounts
+ *
+ * Each entry is a u64 representing the corresponding
+ * physical page mapcount.
+ */
+static ssize_t kpagecount_read(struct file *file, char __user *buf,
+ size_t count, loff_t *ppos)
+{
+ return kpage_read(file, buf, count, ppos, KPAGE_COUNT);
+}
+
static const struct proc_ops kpagecount_proc_ops = {
.proc_flags = PROC_ENTRY_PERMANENT,
.proc_lseek = mem_lseek,
.proc_read = kpagecount_read,
};
-/* /proc/kpageflags - an array exposing page flags
- *
- * Each entry is a u64 representing the corresponding
- * physical page flags.
- */
static inline u64 kpf_copy_bit(u64 kflags, int ubit, int kbit)
{
@@ -220,47 +242,17 @@ u64 stable_page_flags(const struct page *page)
#endif
return u;
-};
+}
+/* /proc/kpageflags - an array exposing page flags
+ *
+ * Each entry is a u64 representing the corresponding
+ * physical page flags.
+ */
static ssize_t kpageflags_read(struct file *file, char __user *buf,
- size_t count, loff_t *ppos)
+ size_t count, loff_t *ppos)
{
- const unsigned long max_dump_pfn = get_max_dump_pfn();
- u64 __user *out = (u64 __user *)buf;
- unsigned long src = *ppos;
- unsigned long pfn;
- ssize_t ret = 0;
-
- pfn = src / KPMSIZE;
- if (src & KPMMASK || count & KPMMASK)
- return -EINVAL;
- if (src >= max_dump_pfn * KPMSIZE)
- return 0;
- count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
-
- while (count > 0) {
- /*
- * TODO: ZONE_DEVICE support requires to identify
- * memmaps that were actually initialized.
- */
- struct page *page = pfn_to_online_page(pfn);
-
- if (put_user(stable_page_flags(page), out)) {
- ret = -EFAULT;
- break;
- }
-
- pfn++;
- out++;
- count -= KPMSIZE;
-
- cond_resched();
- }
-
- *ppos += (char __user *)out - buf;
- if (!ret)
- ret = (char __user *)out - buf;
- return ret;
+ return kpage_read(file, buf, count, ppos, KPAGE_FLAGS);
}
static const struct proc_ops kpageflags_proc_ops = {
@@ -270,54 +262,12 @@ static const struct proc_ops kpageflags_proc_ops = {
};
#ifdef CONFIG_MEMCG
+
static ssize_t kpagecgroup_read(struct file *file, char __user *buf,
- size_t count, loff_t *ppos)
+ size_t count, loff_t *ppos)
{
- const unsigned long max_dump_pfn = get_max_dump_pfn();
- u64 __user *out = (u64 __user *)buf;
- struct page *ppage;
- unsigned long src = *ppos;
- unsigned long pfn;
- ssize_t ret = 0;
- u64 ino;
-
- pfn = src / KPMSIZE;
- if (src & KPMMASK || count & KPMMASK)
- return -EINVAL;
- if (src >= max_dump_pfn * KPMSIZE)
- return 0;
- count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
-
- while (count > 0) {
- /*
- * TODO: ZONE_DEVICE support requires to identify
- * memmaps that were actually initialized.
- */
- ppage = pfn_to_online_page(pfn);
-
- if (ppage)
- ino = page_cgroup_ino(ppage);
- else
- ino = 0;
-
- if (put_user(ino, out)) {
- ret = -EFAULT;
- break;
- }
-
- pfn++;
- out++;
- count -= KPMSIZE;
-
- cond_resched();
- }
-
- *ppos += (char __user *)out - buf;
- if (!ret)
- ret = (char __user *)out - buf;
- return ret;
+ return kpage_read(file, buf, count, ppos, KPAGE_CGROUP);
}
-
static const struct proc_ops kpagecgroup_proc_ops = {
.proc_flags = PROC_ENTRY_PERMANENT,
.proc_lseek = mem_lseek,
--
2.25.1On 17.03.25 11:07, Liu Ye wrote:
> From: Liu Ye <liuye@kylinos.cn>
>
> The function kpageflags_read and kpagecgroup_read is quite similar
> to kpagecount_read. Consider refactoring common code into a helper
> function to reduce code duplication.
>
> Signed-off-by: Liu Ye <liuye@kylinos.cn>
>
> ---
> V2 : Use an enumeration to indicate the operation to be performed
> to avoid passing functions.
> ---
> ---
> fs/proc/page.c | 166 +++++++++++++++++--------------------------------
> 1 file changed, 58 insertions(+), 108 deletions(-)
>
> diff --git a/fs/proc/page.c b/fs/proc/page.c
> index a55f5acefa97..66f454330a87 100644
> --- a/fs/proc/page.c
> +++ b/fs/proc/page.c
> @@ -22,6 +22,14 @@
> #define KPMMASK (KPMSIZE - 1)
> #define KPMBITS (KPMSIZE * BITS_PER_BYTE)
>
> +enum kpage_operation {
> + KPAGE_FLAGS,
> + KPAGE_COUNT,
> +#ifdef CONFIG_MEMCG
> + KPAGE_CGROUP,
> +#endif
> +};
> +
> static inline unsigned long get_max_dump_pfn(void)
> {
> #ifdef CONFIG_SPARSEMEM
> @@ -37,19 +45,17 @@ static inline unsigned long get_max_dump_pfn(void)
> #endif
> }
>
> -/* /proc/kpagecount - an array exposing page mapcounts
> - *
> - * Each entry is a u64 representing the corresponding
> - * physical page mapcount.
> - */
> -static ssize_t kpagecount_read(struct file *file, char __user *buf,
> - size_t count, loff_t *ppos)
> +static ssize_t kpage_read(struct file *file, char __user *buf,
> + size_t count, loff_t *ppos,
> + enum kpage_operation op)
> {
> const unsigned long max_dump_pfn = get_max_dump_pfn();
> u64 __user *out = (u64 __user *)buf;
> + struct page *ppage;
> unsigned long src = *ppos;
> unsigned long pfn;
> ssize_t ret = 0;
> + u64 info;
>
> pfn = src / KPMSIZE;
> if (src & KPMMASK || count & KPMMASK)
> @@ -59,19 +65,29 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
> count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
>
> while (count > 0) {
> - struct page *page;
> - u64 mapcount = 0;
> -
> - /*
> - * TODO: ZONE_DEVICE support requires to identify
> - * memmaps that were actually initialized.
> - */
> - page = pfn_to_online_page(pfn);
> - if (page)
> - mapcount = folio_precise_page_mapcount(page_folio(page),
> - page);
> -
> - if (put_user(mapcount, out)) {
> + ppage = pfn_to_online_page(pfn);
> +
> + if (ppage) {
> + switch (op) {
> + case KPAGE_FLAGS:
> + info = stable_page_flags(ppage);
> + break;
> + case KPAGE_COUNT:
> + info = folio_precise_page_mapcount(page_folio(ppage), ppage);
> + break;
> +#ifdef CONFIG_MEMCG
> + case KPAGE_CGROUP:
> + info = page_cgroup_ino(ppage);
> + break;
> +#endif
In general, LGTM.
I do wonder if we should just get rid of the two "#ifdef CONFIG_MEMCG" by adding
a stub for page_cgroup_ino().
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 57664e2a8fb7b..24248f4dcc971 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -1788,6 +1788,11 @@ static inline void count_objcg_events(struct obj_cgroup *objcg,
{
}
+static inline ino_t page_cgroup_ino(struct page *page)
+{
+ return 0;
+}
+
#endif /* CONFIG_MEMCG */
#if defined(CONFIG_MEMCG) && defined(CONFIG_ZSWAP)
--
Cheers,
David / dhildenb
在 2025/3/17 18:24, David Hildenbrand 写道:
> On 17.03.25 11:07, Liu Ye wrote:
>> From: Liu Ye <liuye@kylinos.cn>
>>
>> The function kpageflags_read and kpagecgroup_read is quite similar
>> to kpagecount_read. Consider refactoring common code into a helper
>> function to reduce code duplication.
>>
>> Signed-off-by: Liu Ye <liuye@kylinos.cn>
>>
>> ---
>> V2 : Use an enumeration to indicate the operation to be performed
>> to avoid passing functions.
>> ---
>> ---
>> fs/proc/page.c | 166 +++++++++++++++++--------------------------------
>> 1 file changed, 58 insertions(+), 108 deletions(-)
>>
>> diff --git a/fs/proc/page.c b/fs/proc/page.c
>> index a55f5acefa97..66f454330a87 100644
>> --- a/fs/proc/page.c
>> +++ b/fs/proc/page.c
>> @@ -22,6 +22,14 @@
>> #define KPMMASK (KPMSIZE - 1)
>> #define KPMBITS (KPMSIZE * BITS_PER_BYTE)
>> +enum kpage_operation {
>> + KPAGE_FLAGS,
>> + KPAGE_COUNT,
>> +#ifdef CONFIG_MEMCG
>> + KPAGE_CGROUP,
>> +#endif
>> +};
>> +
>> static inline unsigned long get_max_dump_pfn(void)
>> {
>> #ifdef CONFIG_SPARSEMEM
>> @@ -37,19 +45,17 @@ static inline unsigned long get_max_dump_pfn(void)
>> #endif
>> }
>> -/* /proc/kpagecount - an array exposing page mapcounts
>> - *
>> - * Each entry is a u64 representing the corresponding
>> - * physical page mapcount.
>> - */
>> -static ssize_t kpagecount_read(struct file *file, char __user *buf,
>> - size_t count, loff_t *ppos)
>> +static ssize_t kpage_read(struct file *file, char __user *buf,
>> + size_t count, loff_t *ppos,
>> + enum kpage_operation op)
>> {
>> const unsigned long max_dump_pfn = get_max_dump_pfn();
>> u64 __user *out = (u64 __user *)buf;
>> + struct page *ppage;
>> unsigned long src = *ppos;
>> unsigned long pfn;
>> ssize_t ret = 0;
>> + u64 info;
>> pfn = src / KPMSIZE;
>> if (src & KPMMASK || count & KPMMASK)
>> @@ -59,19 +65,29 @@ static ssize_t kpagecount_read(struct file *file, char __user *buf,
>> count = min_t(unsigned long, count, (max_dump_pfn * KPMSIZE) - src);
>> while (count > 0) {
>> - struct page *page;
>> - u64 mapcount = 0;
>> -
>> - /*
>> - * TODO: ZONE_DEVICE support requires to identify
>> - * memmaps that were actually initialized.
>> - */
>> - page = pfn_to_online_page(pfn);
>> - if (page)
>> - mapcount = folio_precise_page_mapcount(page_folio(page),
>> - page);
>> -
>> - if (put_user(mapcount, out)) {
>> + ppage = pfn_to_online_page(pfn);
>> +
>> + if (ppage) {
>> + switch (op) {
>> + case KPAGE_FLAGS:
>> + info = stable_page_flags(ppage);
>> + break;
>> + case KPAGE_COUNT:
>> + info = folio_precise_page_mapcount(page_folio(ppage), ppage);
>> + break;
>> +#ifdef CONFIG_MEMCG
>> + case KPAGE_CGROUP:
>> + info = page_cgroup_ino(ppage);
>> + break;
>> +#endif
>
> In general, LGTM.
>
> I do wonder if we should just get rid of the two "#ifdef CONFIG_MEMCG" by adding
> a stub for page_cgroup_ino().
>
> diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
> index 57664e2a8fb7b..24248f4dcc971 100644
> --- a/include/linux/memcontrol.h
> +++ b/include/linux/memcontrol.h
> @@ -1788,6 +1788,11 @@ static inline void count_objcg_events(struct obj_cgroup *objcg,
> {
> }
>
> +static inline ino_t page_cgroup_ino(struct page *page)
> +{
> + return 0;
> +}
> +
> #endif /* CONFIG_MEMCG */
>
> #if defined(CONFIG_MEMCG) && defined(CONFIG_ZSWAP)
>
>
Agreed. I’ll add a stub for page_cgroup_ino() and remove the #ifdef CONFIG_MEMCG.
like this.
diff --git a/fs/proc/page.c b/fs/proc/page.c
index 66f454330a87..cbadbf9568a1 100644
--- a/fs/proc/page.c
+++ b/fs/proc/page.c
@@ -25,9 +25,7 @@
enum kpage_operation {
KPAGE_FLAGS,
KPAGE_COUNT,
-#ifdef CONFIG_MEMCG
KPAGE_CGROUP,
-#endif
};
static inline unsigned long get_max_dump_pfn(void)
@@ -75,11 +73,9 @@ static ssize_t kpage_read(struct file *file, char __user *buf,
case KPAGE_COUNT:
info = folio_precise_page_mapcount(page_folio(ppage), ppage);
break;
-#ifdef CONFIG_MEMCG
case KPAGE_CGROUP:
info = page_cgroup_ino(ppage);
break;
-#endif
default:
info = 0;
break;
@@ -262,7 +258,6 @@ static const struct proc_ops kpageflags_proc_ops = {
};
#ifdef CONFIG_MEMCG
-
static ssize_t kpagecgroup_read(struct file *file, char __user *buf,
size_t count, loff_t *ppos)
{
diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 6e74b8254d9b..e806e2ebf5b8 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -1794,6 +1794,10 @@ static inline void count_objcg_events(struct obj_cgroup *objcg,
{
}
+static inline ino_t page_cgroup_ino(struct page *page)
+{
+ return 0;
+}
#endif /* CONFIG_MEMCG */
#if defined(CONFIG_MEMCG) && defined(CONFIG_ZSWAP)
© 2016 - 2025 Red Hat, Inc.