Some applications that run on HPC clusters are designed around the
availability of RAM and the overcommit ratio is fine tuned to get the
maximum usage of memory without swapping. With growing memory, the
1%-of-all-RAM grain provided by overcommit_ratio has become too coarse
for these workload (on a 2TB machine it represents no less than
20GB).
This patch adds the new overcommit_kbytes sysctl variable that allow a
much finer grain.
Signed-of-by: Jerome Marchand <jmarchan@xxxxxxxxxx>
---
Documentation/sysctl/vm.txt | 12 ++++++++++++
Documentation/vm/overcommit-accounting | 7 ++++---
include/linux/mm.h | 5 +++++
include/linux/mman.h | 13 +++++++++++--
kernel/sysctl.c | 10 +++++++++-
mm/mmap.c | 25 +++++++++++++++++++++++++
mm/nommu.c | 1 +
7 files changed, 67 insertions(+), 6 deletions(-)
diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 36ecc26..a23aea1 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -47,6 +47,7 @@ Currently, these files are in /proc/sys/vm:
- numa_zonelist_order
- oom_dump_tasks
- oom_kill_allocating_task
+- overcommit_kbytes
- overcommit_memory
- overcommit_ratio
- page-cluster
@@ -561,6 +562,17 @@ The default value is 0.
==============================================================
+overcommit_kbytes:
+
+When overcommit_memory is set to 2, the committed address space is not
+permitted to exceed swap plus this amount of physical RAM. See below.
+
+Note: overcommit_kbytes is the counterpart of overcommit_ratio. Only one
+of them may be specified at a time. Setting one disable the other (which
+then appears as 0 when read).
+
+==============================================================
+
overcommit_memory:
This value contains a flag that enables memory overcommitment.
diff --git a/Documentation/vm/overcommit-accounting b/Documentation/vm/overcommit-accounting
index 8eaa2fc..cbfaaa6 100644
--- a/Documentation/vm/overcommit-accounting
+++ b/Documentation/vm/overcommit-accounting
@@ -14,8 +14,8 @@ The Linux kernel supports the following overcommit handling modes
2 - Don't overcommit. The total address space commit
for the system is not permitted to exceed swap + a
- configurable percentage (default is 50) of physical RAM.
- Depending on the percentage you use, in most situations
+ configurable amount (default is 50%) of physical RAM.
+ Depending on the amount you use, in most situations
this means a process will not be killed while accessing
pages but will receive errors on memory allocation as
appropriate.
@@ -26,7 +26,8 @@ The Linux kernel supports the following overcommit handling modes
The overcommit policy is set via the sysctl `vm.overcommit_memory'.
-The overcommit percentage is set via `vm.overcommit_ratio'.
+The overcommit amount can be set via `vm.overcommit_ratio' (percentage)
+or `vm.overcommit_kbytes' (absolute value).
The current overcommit limit and amount committed are viewable in
/proc/meminfo as CommitLimit and Committed_AS respectively.
diff --git a/include/linux/mm.h b/include/linux/mm.h
index f022460..15f2b0c 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -53,6 +53,11 @@ extern int sysctl_legacy_va_layout;
extern unsigned long sysctl_user_reserve_kbytes;
extern unsigned long sysctl_admin_reserve_kbytes;
+extern int overcommit_ratio_handler(struct ctl_table *, int, void __user *,
+ size_t *, loff_t *);
+extern int overcommit_kbytes_handler(struct ctl_table *, int, void __user *,
+ size_t *, loff_t *);
+
#define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))
/* to align the pointer to the (next) page boundary */
diff --git a/include/linux/mman.h b/include/linux/mman.h
index d622d34..debd0f9 100644
--- a/include/linux/mman.h
+++ b/include/linux/mman.h
@@ -12,6 +12,7 @@
extern int sysctl_overcommit_memory;
extern int sysctl_overcommit_ratio;
+extern unsigned long sysctl_overcommit_kbytes;
extern struct percpu_counter vm_committed_as;
#ifdef CONFIG_SMP
@@ -96,7 +97,15 @@ calc_vm_flag_bits(unsigned long flags)
*/
static inline unsigned long vm_commit_limit()
{
- return ((totalram_pages - hugetlb_total_pages())
- * sysctl_overcommit_ratio / 100) + total_swap_pages;
+ unsigned long allowed;
+
+ if (sysctl_overcommit_kbytes)
+ allowed = sysctl_overcommit_kbytes >> (PAGE_SHIFT - 10);
+ else
+ allowed = ((totalram_pages - hugetlb_total_pages())
+ * sysctl_overcommit_ratio / 100);
+ allowed += total_swap_pages;
+
+ return allowed;
}
#endif /* _LINUX_MMAN_H */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 07f6fc4..e1968a4 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -97,6 +97,7 @@
/* External variables not in a header file. */
extern int sysctl_overcommit_memory;
extern int sysctl_overcommit_ratio;
+extern unsigned long sysctl_overcommit_kbytes;
extern int max_threads;
extern int suid_dumpable;
#ifdef CONFIG_COREDUMP
@@ -1119,7 +1120,14 @@ static struct ctl_table vm_table[] = {
.data = &sysctl_overcommit_ratio,
.maxlen = sizeof(sysctl_overcommit_ratio),
.mode = 0644,
- .proc_handler = proc_dointvec,
+ .proc_handler = overcommit_ratio_handler,
+ },
+ {
+ .procname = "overcommit_kbytes",
+ .data = &sysctl_overcommit_kbytes,
+ .maxlen = sizeof(sysctl_overcommit_kbytes),
+ .mode = 0644,
+ .proc_handler = overcommit_kbytes_handler,
},
{
.procname = "page-cluster",
diff --git a/mm/mmap.c b/mm/mmap.c
index 06c98f8..c4a09a5 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -86,6 +86,7 @@ EXPORT_SYMBOL(vm_get_page_prot);
int sysctl_overcommit_memory __read_mostly = OVERCOMMIT_GUESS; /* heuristic overcommit */
int sysctl_overcommit_ratio __read_mostly = 50; /* default is 50% */
+unsigned long sysctl_overcommit_kbytes __read_mostly = 0;
int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT;
unsigned long sysctl_user_reserve_kbytes __read_mostly = 1UL << 17; /* 128MB */
unsigned long sysctl_admin_reserve_kbytes __read_mostly = 1UL << 13; /* 8MB */
@@ -95,6 +96,30 @@ unsigned long sysctl_admin_reserve_kbytes __read_mostly = 1UL << 13; /* 8MB */
*/
struct percpu_counter vm_committed_as ____cacheline_aligned_in_smp;
+int overcommit_ratio_handler(struct ctl_table *table, int write,
+ void __user *buffer, size_t *lenp,
+ loff_t *ppos)
+{
+ int ret;
+
+ ret = proc_dointvec(table, write, buffer, lenp, ppos);
+ if (ret == 0 && write)
+ sysctl_overcommit_kbytes = 0;
+ return ret;
+}
+
+int overcommit_kbytes_handler(struct ctl_table *table, int write,
+ void __user *buffer, size_t *lenp,
+ loff_t *ppos)
+{
+ int ret;
+
+ ret = proc_doulongvec_minmax(table, write, buffer, lenp, ppos);
+ if (ret == 0 && write)
+ sysctl_overcommit_ratio = 0;
+ return ret;
+}
+
/*
* The global memory commitment made in the system can be a metric
* that can be used to drive ballooning decisions when Linux is hosted
diff --git a/mm/nommu.c b/mm/nommu.c
index d8a957b..3c4216a 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -60,6 +60,7 @@ unsigned long highest_memmap_pfn;
struct percpu_counter vm_committed_as;
int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */
int sysctl_overcommit_ratio = 50; /* default is 50% */
+unsigned long sysctl_overcommit_kbytes __read_mostly = 0;
int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT;
int sysctl_nr_trim_pages = CONFIG_NOMMU_INITIAL_TRIM_EXCESS;
unsigned long sysctl_user_reserve_kbytes __read_mostly = 1UL << 17; /* 128MB */
--
1.7.7.6
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>