Skip to contentSkip to conversation

Add per cpuset meminfo

PATCH In review Ying Han 2 participants 4 messages Dec 10, 2010

Subsystems (2)
lore
Share & tools
lore

Ying Han

author
#0 lore raw

cover Add per cpuset meminfo

Export per cpuset meminfo through cpuset.meminfo file. This is easier than user program to aggregate it across each nodes in nodemask.

Ying Han (2):
Add hugetlb_report_nodemask_meminfo()
Add per cpuset meminfo

 include/linux/hugetlb.h |    3 +
 kernel/cpuset.c         |  118 +++++++++++++++++++++++++++++++++++++++++++++++
 mm/hugetlb.c            |   21 ++++++++
 3 files changed, 142 insertions(+), 0 deletions(-)
Signature · 8 lines
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
author
#1 lore raw

PATCH 1/2 Add hugetlb_report_nodemask_meminfo()

It is useful to accumulate hugetlb statistics on nodemask. This is used on the following patch also which exporting per-cpuset meminfo.

---
 include/linux/hugetlb.h |    3 +++
 mm/hugetlb.c            |   21 +++++++++++++++++++++
 2 files changed, 24 insertions(+), 0 deletions(-)
include/linux/hugetlb.h +3 -0
index 943c76b..5e95672 100644
@@ -10,6 +10,7 @@ struct user_struct;
 #ifdef CONFIG_HUGETLB_PAGE
 
 #include <linux/mempolicy.h>
+#include <linux/nodemask.h>
 #include <linux/shm.h>
 #include <asm/tlbflush.h>
 
@@ -36,6 +37,7 @@ void __unmap_hugepage_range(struct vm_area_struct *,
 int hugetlb_prefault(struct address_space *, struct vm_area_struct *);
 void hugetlb_report_meminfo(struct seq_file *);
 int hugetlb_report_node_meminfo(int, char *);
+void hugetlb_report_nodemask_meminfo(const nodemask_t *mask, struct seq_file *);
 unsigned long hugetlb_total_pages(void);
 int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, unsigned int flags);
@@ -93,6 +95,7 @@ static inline void hugetlb_report_meminfo(struct seq_file *m)
 {
 }
 #define hugetlb_report_node_meminfo(n, buf)	0
+#define hugetlb_report_nodemask_meminfo(mask, seq_file)
 #define follow_huge_pmd(mm, addr, pmd, write)	NULL
 #define follow_huge_pud(mm, addr, pud, write)	NULL
 #define prepare_hugepage_range(file, addr, len)	(-EINVAL)
mm/hugetlb.c +21 -0
index c4a3558..27961eb 100644
@@ -2029,6 +2029,27 @@ int hugetlb_report_node_meminfo(int nid, char *buf)
 		nid, h->surplus_huge_pages_node[nid]);
 }
 
+void hugetlb_report_nodemask_meminfo(const nodemask_t *mask, struct seq_file *m)
+{
+	unsigned long total = 0;
+	unsigned long free = 0;
+	unsigned long surp = 0;
+	struct hstate *h = &default_hstate;
+	int nid;
+
+	for_each_node_mask(nid, *mask) {
+		total += h->nr_huge_pages_node[nid];
+		free += h->free_huge_pages_node[nid];
+		surp += h->surplus_huge_pages_node[nid];
+	}
+
+	seq_printf(m,
+		"HugePages_Total: %5lu\n"
+		"HugePages_Free:  %5lu\n"
+		"HugePages_Surp:  %5lu\n",
+		total, free, surp);
+}
+
 /* Return the number pages of memory we physically have, in PAGE_SIZE units. */
 unsigned long hugetlb_total_pages(void)
 {
Signature · 8 lines
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
author
#2 lore raw

PATCH 2/2 Add per cpuset meminfo

In reply to Ying Han

This exports accumulated memory information through cpuset.meminfo for each cpuset. This is a useful extention which userspace program doesn't have to perform the aggregation itself.

---
 kernel/cpuset.c |  118 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 118 insertions(+), 0 deletions(-)
kernel/cpuset.c +118 -0
index 51b143e..815c375 100644
@@ -53,6 +53,7 @@
 #include <linux/time.h>
 #include <linux/backing-dev.h>
 #include <linux/sort.h>
+#include <linux/hugetlb.h>
 
 #include <asm/uaccess.h>
 #include <asm/atomic.h>
@@ -1720,6 +1721,118 @@ static s64 cpuset_read_s64(struct cgroup *cont, struct cftype *cft)
 	return 0;
 }
 
+#define K(x) ((x) << (PAGE_SHIFT - 10))
+static int cpuset_read_meminfo(struct cgroup *cont, struct cftype *cft,
+			       struct seq_file *m)
+{
+	struct cpuset *cs = cgroup_cs(cont);
+	struct sysinfo si_accum;
+	unsigned long *node_info;
+	int nid;
+	NODEMASK_ALLOC(nodemask_t, mask, GFP_KERNEL);
+
+	if (mask == NULL)
+		return -ENOMEM;
+	node_info = kzalloc(sizeof(*node_info) * NR_VM_ZONE_STAT_ITEMS,
+			    GFP_KERNEL);
+	if (node_info == NULL) {
+		NODEMASK_FREE(mask);
+		return -ENOMEM;
+	}
+
+	mutex_lock(&callback_mutex);
+	*mask = cs->mems_allowed;
+	mutex_unlock(&callback_mutex);
+
+	memset(&si_accum, 0, sizeof(si_accum));
+	for_each_node_mask(nid, *mask) {
+		int i;
+		struct sysinfo si_node;
+
+		si_meminfo_node(&si_node, nid);
+		si_accum.totalram += si_node.totalram;
+		si_accum.freeram += si_node.freeram;
+#ifdef CONFIG_HIGHMEM
+		si_accum.totalhigh += si_node.totalhigh;
+		si_accum.freehigh += si_node.freehigh;
+#endif
+
+		for (i = 0; i < NR_VM_ZONE_STAT_ITEMS; i++)
+			node_info[i] += node_page_state(nid, i);
+	}
+
+	seq_printf(m,
+		   "MemTotal:       %8lu kB\n"
+		   "MemFree:        %8lu kB\n"
+		   "MemUsed:        %8lu kB\n"
+		   "Active:         %8lu kB\n"
+		   "Inactive:       %8lu kB\n"
+		   "Active(anon):   %8lu kB\n"
+		   "Inactive(anon): %8lu kB\n"
+		   "Active(file):   %8lu kB\n"
+		   "Inactive(file): %8lu kB\n"
+		   "Unevictable:    %8lu kB\n"
+		   "Mlock:          %8lu kB\n"
+#ifdef CONFIG_HIGHMEM
+		   "HighTotal:      %8lu kB\n"
+		   "HighFree:       %8lu kB\n"
+		   "LowTotal:       %8lu kB\n"
+		   "LowFree:        %8lu kB\n"
+#endif
+		   "Dirty:          %8lu kB\n"
+		   "Writeback:      %8lu kB\n"
+		   "FilePages:      %8lu kB\n"
+		   "Mapped:         %8lu kB\n"
+		   "AnonPages:      %8lu kB\n"
+		   "Shmem:          %8lu kB\n"
+		   "KernelStack:    %8lu kB\n"
+		   "PageTables:     %8lu kB\n"
+		   "NFS_Unstable:   %8lu kB\n"
+		   "Bounce:         %8lu kB\n"
+		   "WritebackTmp:   %8lu kB\n"
+		   "Slab:           %8lu kB\n"
+		   "SReclaimable:   %8lu kB\n"
+		   "SUnreclaim:     %8lu kB\n",
+		   K(si_accum.totalram),
+		   K(si_accum.freeram),
+		   K(si_accum.totalram - si_accum.freeram),
+		   K(node_info[NR_ACTIVE_ANON] + node_info[NR_ACTIVE_FILE]),
+		   K(node_info[NR_INACTIVE_ANON] +
+			node_info[NR_INACTIVE_FILE]),
+		   K(node_info[NR_ACTIVE_ANON]),
+		   K(node_info[NR_INACTIVE_ANON]),
+		   K(node_info[NR_ACTIVE_FILE]),
+		   K(node_info[NR_INACTIVE_FILE]),
+		   K(node_info[NR_UNEVICTABLE]),
+		   K(node_info[NR_MLOCK]),
+#ifdef CONFIG_HIGHMEM
+		   K(si_accum.totalhigh),
+		   K(si_accum.freehigh),
+		   K(si_accum.totalram - si_accum.totalhigh),
+		   K(si_accum.freeram - si_accum.freehigh),
+#endif
+		   K(node_info[NR_FILE_DIRTY]),
+		   K(node_info[NR_WRITEBACK]),
+		   K(node_info[NR_FILE_PAGES]),
+		   K(node_info[NR_FILE_MAPPED]),
+		   K(node_info[NR_ANON_PAGES]),
+		   K(node_info[NR_SHMEM]),
+			node_info[NR_KERNEL_STACK] * THREAD_SIZE / 1024,
+		   K(node_info[NR_PAGETABLE]),
+		   K(node_info[NR_UNSTABLE_NFS]),
+		   K(node_info[NR_BOUNCE]),
+		   K(node_info[NR_WRITEBACK_TEMP]),
+		   K(node_info[NR_SLAB_RECLAIMABLE] +
+			node_info[NR_SLAB_UNRECLAIMABLE]),
+		   K(node_info[NR_SLAB_RECLAIMABLE]),
+		   K(node_info[NR_SLAB_UNRECLAIMABLE]));
+	hugetlb_report_nodemask_meminfo(mask, m);
+
+	kfree(node_info);
+	NODEMASK_FREE(mask);
+
+	return 0;
+}
 
 /*
  * for the common functions, 'private' gives the type of file
@@ -1805,6 +1918,11 @@ static struct cftype files[] = {
 		.write_u64 = cpuset_write_u64,
 		.private = FILE_SPREAD_SLAB,
 	},
+
+	{
+		.name = "meminfo",
+		.read_seq_string = cpuset_read_meminfo,
+	},
 };
 
 static struct cftype cft_memory_pressure_enabled = {
Signature · 8 lines
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
#3 lore raw
In reply to Ying Han
Quoted 2 lines

Export per cpuset meminfo through cpuset.meminfo file. This is easier than user program to aggregate it across each nodes in nodemask.

Now userland folks calculate it by userland tools by using /sys/device/node/meminfo. So, The description should explain why do you want to avoid userland calculation. Avoid annoying? Or you've found some corner case?

Also please avoid cut-n-paste code duplication. please try unify with other meminfo code.

Quoted 19 lines

Ying Han (2):
Add hugetlb_report_nodemask_meminfo()
Add per cpuset meminfo

 include/linux/hugetlb.h |    3 +
 kernel/cpuset.c         |  118 +++++++++++++++++++++++++++++++++++++++++++++++
 mm/hugetlb.c            |   21 ++++++++
 3 files changed, 142 insertions(+), 0 deletions(-)
Signature · 8 lines
1.7.3.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>