The patch titled
mm: quicklist shouldn't be proportional to number of CPUs
has been removed from the -mm tree. Its filename was
mm-quicklist-shouldnt-be-proportional-to-number-of-cpus.patch
This patch was dropped because an updated version will be merged
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: mm: quicklist shouldn't be proportional to number of CPUs
From: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
When a test program which does task migration runs, my 8GB box spends
800MB of memory for quicklist. This is not memory leak but doesn't seem
good.
% cat /proc/meminfo
MemTotal: 7701568 kB
MemFree: 4724672 kB
(snip)
Quicklists: 844800 kB
because
- My machine spec is
number of numa node: 2
number of cpus: 8 (4CPU x2 node)
total mem: 8GB (4GB x2 node)
free mem: about 5GB
- Maximum quicklist usage is here
Number of CPUs per node 2 4 8 16
============================== ====================
QList_max / (Free + QList_max) 5.8% 16% 30% 48%
- Then, 4.7GB x 16% ~= 880MB.
So, Quicklist can use 800MB.
So, if following spec machine run that program
CPUs: 64 (8cpu x 8node)
Mem: 1TB (128GB x8node)
Then, quicklist can waste 300GB (= 1TB x 30%). It is too large.
So, I don't like cache policies which is proportional to # of cpus.
My patch changes the number of caches
from:
per-cpu-cache-amount = memory_on_node / 16
to
per-cpu-cache-amount = memory_on_node / 16 / number_of_cpus_on_node.
I think this is reasonable. but even if this patch is applied, quicklist
can cache tons of memory on big machine.
(Although its patch applied, quicklist can waste 64GB on 1TB server (= 1TB
/ 16), it is still too much??)
test program is below.
--------------------------------------------------------------------------------
#define _GNU_SOURCE
#include <stdio.h>
#include <errno.h>
#include <stdlib.h>
#include <string.h>
#include <sched.h>
#include <unistd.h>
#include <sys/mman.h>
#include <sys/wait.h>
#define BUFFSIZE 512
int max_cpu(void) /* get max number of logical cpus from /proc/cpuinfo */
{
FILE *fd;
char *ret, buffer[BUFFSIZE];
int cpu = 1;
fd = fopen("/proc/cpuinfo", "r");
if (fd == NULL) {
perror("fopen(/proc/cpuinfo)");
exit(EXIT_FAILURE);
}
while (1) {
ret = fgets(buffer, BUFFSIZE, fd);
if (ret == NULL)
break;
if (!strncmp(buffer, "processor", 9))
cpu = atoi(strchr(buffer, ':') + 2);
}
fclose(fd);
return cpu;
}
void cpu_bind(int cpu) /* bind current process to one cpu */
{
cpu_set_t mask;
int ret;
CPU_ZERO(&mask);
CPU_SET(cpu, &mask);
ret = sched_setaffinity(0, sizeof(mask), &mask);
if (ret == -1) {
perror("sched_setaffinity()");
exit(EXIT_FAILURE);
}
sched_yield(); /* not necessary */
}
#define MMAP_SIZE (10 * 1024 * 1024) /* 10 MB */
#define FORK_INTERVAL 1 /* 1 second */
main(int argc, char *argv[])
{
int cpu_max, nextcpu;
long pagesize;
pid_t pid;
/* set max number of logical cpu */
if (argc > 1)
cpu_max = atoi(argv[1]) - 1;
else
cpu_max = max_cpu();
/* get the page size */
pagesize = sysconf(_SC_PAGESIZE);
if (pagesize == -1) {
perror("sysconf(_SC_PAGESIZE)");
exit(EXIT_FAILURE);
}
/* prepare parent process */
cpu_bind(0);
nextcpu = cpu_max;
loop:
/* select destination cpu for child process by round-robin rule */
if (++nextcpu > cpu_max)
nextcpu = 1;
pid = fork();
if (pid == 0) { /* child action */
char *p;
int i;
/* consume page tables */
p = mmap(0, MMAP_SIZE, PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, 0, 0);
i = MMAP_SIZE / pagesize;
while (i-- > 0) {
*p = 1;
p += pagesize;
}
/* move to other cpu */
cpu_bind(nextcpu);
/*
printf("a child moved to cpu%d after mmap().\n", nextcpu);
fflush(stdout);
*/
/* back page tables to pgtable_quicklist */
exit(0);
} else if (pid > 0) { /* parent action */
sleep(FORK_INTERVAL);
waitpid(pid, NULL, WNOHANG);
}
goto loop;
}
[akpm@xxxxxxxxxxxxxxxxxxxx: fix build on sparc64]
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Cc: Christoph Lameter <cl@xxxxxxxxxxxxxxxxxxxx>
Cc: Mike Travis <travis@xxxxxxx>
Cc: <stable@xxxxxxxxxx> [2.6.25.x, 2.6.26.x]
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/quicklist.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff -puN mm/quicklist.c~mm-quicklist-shouldnt-be-proportional-to-number-of-cpus mm/quicklist.c
--- a/mm/quicklist.c~mm-quicklist-shouldnt-be-proportional-to-number-of-cpus
+++ a/mm/quicklist.c
@@ -26,7 +26,9 @@ DEFINE_PER_CPU(struct quicklist, quickli
static unsigned long max_pages(unsigned long min_pages)
{
unsigned long node_free_pages, max;
- struct zone *zones = NODE_DATA(numa_node_id())->node_zones;
+ int node = numa_node_id();
+ struct zone *zones = NODE_DATA(node)->node_zones;
+ cpumask_t node_cpumask;
node_free_pages =
#ifdef CONFIG_ZONE_DMA
@@ -38,6 +40,10 @@ static unsigned long max_pages(unsigned
zone_page_state(&zones[ZONE_NORMAL], NR_FREE_PAGES);
max = node_free_pages / FRACTION_OF_NODE_MEM;
+
+ node_cpumask = node_to_cpumask(node);
+ max /= cpus_weight_nr(node_cpumask);
+
return max(max, min_pages);
}
_
Patches currently in -mm which might be from kosaki.motohiro@xxxxxxxxxxxxxx are
mm-quicklist-shouldnt-be-proportional-to-number-of-cpus.patch
linux-next.patch
vmscan-use-an-indexed-array-for-lru-variables.patch
swap-use-an-array-for-the-lru-pagevecs.patch
vmscan-split-lru-lists-into-anon-file-sets.patch
vmscan-second-chance-replacement-for-anonymous-pages.patch
unevictable-lru-infrastructure.patch
unevictable-lru-infrastructure-nommu-fix.patch
unevictable-lru-infrastructure-remember-pages-active-state.patch
unevictable-lru-infrastructure-defer-vm-event-counting.patch
unevictable-infrastructure-lru-add-event-counting-with-statistics.patch
unevictable-lru-page-statistics.patch
shm_locked-pages-are-unevictable.patch
shm_locked-pages-are-unevictable-add-event-counts-to-list-scan.patch
mlock-mlocked-pages-are-unevictable.patch
mlock-mlocked-pages-are-unevictable-fix.patch
doc-unevictable-lru-and-mlocked-pages-documentation-update-2.patch
mmap-handle-mlocked-pages-during-map-remap-unmap.patch
mmap-handle-mlocked-pages-during-map-remap-unmap-mlock-fix-__mlock_vma_pages_range-comment-block.patch
mmap-handle-mlocked-pages-during-map-remap-unmap-mlock-backout-locked_vm-adjustment-during-mmap.patch
mmap-handle-mlocked-pages-during-map-remap-unmap-mlock-resubmit-locked_vm-adjustment-as-separate-patch.patch
mmap-handle-mlocked-pages-during-map-remap-unmap-mlock-resubmit-locked_vm-adjustment-as-separate-patch-fix.patch
mmap-handle-mlocked-pages-during-map-remap-unmap-mlock-fix-return-value-for-munmap-mlock-vma-race.patch
mmap-handle-mlocked-pages-during-map-remap-unmap-mlock-update-locked_vm-on-munmap-of-mlocked-region.patch
vmstat-mlocked-pages-statistics.patch
vmstat-mlocked-pages-statistics-mlocked-pages-add-event-counting-with-statistics.patch
swap-cull-unevictable-pages-in-fault-path.patch
vmscan-unevictable-lru-scan-sysctl.patch
vmscam-kill-unused-lru-functions.patch
mlock-revert-mainline-handling-of-mlock-error-return.patch
mlock-make-mlock-error-return-posixly-correct.patch
mlock-make-mlock-error-return-posixly-correct-fix.patch
mm-unlockless-reclaim.patch
mm-more-likely-reclaim-madv_sequential-mappings.patch
make-mm-rmapc-anon_vma_cachep-static.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
