aboutsummaryrefslogtreecommitdiffstats
path: root/kernel
diff options
context:
space:
mode:
authorHugh Dickins <hugh@veritas.com>2005-10-29 21:16:18 -0400
committerLinus Torvalds <torvalds@g5.osdl.org>2005-10-30 00:40:39 -0400
commit365e9c87a982c03d0af3886e29d877f581b59611 (patch)
treed06c1918ca9fe6677d7e4e869555e095004274f7 /kernel
parent861f2fb8e796022b4928cab9c74fca6681a1c557 (diff)
[PATCH] mm: update_hiwaters just in time
update_mem_hiwater has attracted various criticisms, in particular from those concerned with mm scalability. Originally it was called whenever rss or total_vm got raised. Then many of those callsites were replaced by a timer tick call from account_system_time. Now Frank van Maarseveen reports that to be found inadequate. How about this? Works for Frank. Replace update_mem_hiwater, a poor combination of two unrelated ops, by macros update_hiwater_rss and update_hiwater_vm. Don't attempt to keep mm->hiwater_rss up to date at timer tick, nor every time we raise rss (usually by 1): those are hot paths. Do the opposite, update only when about to lower rss (usually by many), or just before final accounting in do_exit. Handle mm->hiwater_vm in the same way, though it's much less of an issue. Demand that whoever collects these hiwater statistics do the work of taking the maximum with rss or total_vm. And there has been no collector of these hiwater statistics in the tree. The new convention needs an example, so match Frank's usage by adding a VmPeak line above VmSize to /proc/<pid>/status, and also a VmHWM line above VmRSS (High-Water-Mark or High-Water-Memory). There was a particular anomaly during mremap move, that hiwater_vm might be captured too high. A fleeting such anomaly remains, but it's quickly corrected now, whereas before it would stick. What locking? None: if the app is racy then these statistics will be racy, it's not worth any overhead to make them exact. But whenever it suits, hiwater_vm is updated under exclusive mmap_sem, and hiwater_rss under page_table_lock (for now) or with preemption disabled (later on): without going to any trouble, minimize the time between reading current values and updating, to minimize those occasions when a racing thread bumps a count up and back down in between. Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'kernel')
-rw-r--r--kernel/exit.c5
-rw-r--r--kernel/sched.c2
2 files changed, 4 insertions, 3 deletions
diff --git a/kernel/exit.c b/kernel/exit.c
index 3b25b182d2be..79f52b85d6ed 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -839,7 +839,10 @@ fastcall NORET_TYPE void do_exit(long code)
839 preempt_count()); 839 preempt_count());
840 840
841 acct_update_integrals(tsk); 841 acct_update_integrals(tsk);
842 update_mem_hiwater(tsk); 842 if (tsk->mm) {
843 update_hiwater_rss(tsk->mm);
844 update_hiwater_vm(tsk->mm);
845 }
843 group_dead = atomic_dec_and_test(&tsk->signal->live); 846 group_dead = atomic_dec_and_test(&tsk->signal->live);
844 if (group_dead) { 847 if (group_dead) {
845 del_timer_sync(&tsk->signal->real_timer); 848 del_timer_sync(&tsk->signal->real_timer);
diff --git a/kernel/sched.c b/kernel/sched.c
index 1e5cafdf4e27..4f26c544d02c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -2511,8 +2511,6 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
2511 cpustat->idle = cputime64_add(cpustat->idle, tmp); 2511 cpustat->idle = cputime64_add(cpustat->idle, tmp);
2512 /* Account for system time used */ 2512 /* Account for system time used */
2513 acct_update_integrals(p); 2513 acct_update_integrals(p);
2514 /* Update rss highwater mark */
2515 update_mem_hiwater(p);
2516} 2514}
2517 2515
2518/* 2516/*