On Thu, 26 May 2011 21:33:32 -0700 Ying Han <yinghan@xxxxxxxxxx> wrote: > On Thu, May 26, 2011 at 7:16 PM, KAMEZAWA Hiroyuki > <kamezawa.hiroyu@xxxxxxxxxxxxxx> wrote: > > On Thu, 26 May 2011 18:49:26 -0700 > > Ying Han <yinghan@xxxxxxxxxx> wrote: > > > >> On Wed, May 25, 2011 at 10:10 PM, KAMEZAWA Hiroyuki > >> <kamezawa.hiroyu@xxxxxxxxxxxxxx> wrote: > >> > > >> > It's now merge window...I just dump my patch queue to hear other's idea. > >> > I wonder I should wait until dirty_ratio for memcg is queued to mmotm... > >> > I'll be busy with LinuxCon Japan etc...in the next week. > >> > > >> > This patch is onto mmotm-May-11 + some patches queued in mmotm, as numa_stat. > >> > > >> > This is a patch for memcg to keep margin to the limit in background. > >> > By keeping some margin to the limit in background, application can > >> > avoid foreground memory reclaim at charge() and this will help latency. > >> > > >> > Main changes from v2 is. > >> > Â- use SCHED_IDLE. > >> > Â- removed most of heuristic codes. Now, code is very simple. > >> > > >> > By using SCHED_IDLE, async memory reclaim can only consume 0.3%? of cpu > >> > if the system is truely busy but can use much CPU if the cpu is idle. > >> > Because my purpose is for reducing latency without affecting other running > >> > applications, SCHED_IDLE fits this work. > >> > > >> > If application need to stop by some I/O or event, background memory reclaim > >> > will cull memory while the system is idle. > >> > > >> > Perforemce: > >> > ÂRunning an httpd (apache) under 300M limit. And access 600MB working set > >> > Âwith normalized distribution access by apatch-bench. > >> > Âapatch bench's concurrency was 4 and did 40960 accesses. > >> > > >> > Without async reclaim: > >> > Connection Times (ms) > >> >       Âmin Âmean[+/-sd] median  max > >> > Connect:    Â0  Â0  0.0   Â0    2 > >> > Processing:  Â30  37 Â28.3   32  Â1793 > >> > Waiting:    28  35 Â25.5   31  Â1792 > >> > Total:     30  37 Â28.4   32  Â1793 > >> > > >> > Percentage of the requests served within a certain time (ms) > >> > Â50%   32 > >> > Â66%   32 > >> > Â75%   33 > >> > Â80%   34 > >> > Â90%   39 > >> > Â95%   60 > >> > Â98%  Â100 > >> > Â99%  Â133 > >> > Â100%  1793 (longest request) > >> > > >> > With async reclaim: > >> > Connection Times (ms) > >> >       Âmin Âmean[+/-sd] median  max > >> > Connect:    Â0  Â0  0.0   Â0    2 > >> > Processing:  Â30  35 Â12.3   32   678 > >> > Waiting:    28  34 Â12.0   31   658 > >> > Total:     30  35 Â12.3   32   678 > >> > > >> > Percentage of the requests served within a certain time (ms) > >> > Â50%   32 > >> > Â66%   32 > >> > Â75%   33 > >> > Â80%   34 > >> > Â90%   39 > >> > Â95%   49 > >> > Â98%   71 > >> > Â99%   86 > >> > Â100%  Â678 (longest request) > >> > > >> > > >> > It seems latency is stabilized by hiding memory reclaim. > >> > > >> > The score for memory reclaim was following. > >> > See patch 10 for meaning of each member. > >> > > >> > == without async reclaim == > >> > recent_scan_success_ratio 44 > >> > limit_scan_pages 388463 > >> > limit_freed_pages 162238 > >> > limit_elapsed_ns 13852159231 > >> > soft_scan_pages 0 > >> > soft_freed_pages 0 > >> > soft_elapsed_ns 0 > >> > margin_scan_pages 0 > >> > margin_freed_pages 0 > >> > margin_elapsed_ns 0 > >> > > >> > == with async reclaim == > >> > recent_scan_success_ratio 6 > >> > limit_scan_pages 0 > >> > limit_freed_pages 0 > >> > limit_elapsed_ns 0 > >> > soft_scan_pages 0 > >> > soft_freed_pages 0 > >> > soft_elapsed_ns 0 > >> > margin_scan_pages 1295556 > >> > margin_freed_pages 122450 > >> > margin_elapsed_ns 644881521 > >> > > >> > > >> > For this case, SCHED_IDLE workqueue can reclaim enough memory to the httpd. > >> > > >> > I may need to dig why scan_success_ratio is far different in the both case. > >> > I guess the difference of epalsed_ns is because several threads enter > >> > memory reclaim when async reclaim doesn't run. But may not... > >> > > >> > >> > >> Hmm.. I noticed a very strange behavior on a simple test w/ the patch set. > >> > >> Test: > >> I created a 4g memcg and start doing cat. Then the memcg being OOM > >> killed as soon as it reaches its hard_limit. We shouldn't hit OOM even > >> w/o async-reclaim. > >> > >> Again, I will read through the patch. But like to post the test result first. > >> > >> $ echo $$ >/dev/cgroup/memory/A/tasks > >> $ cat /dev/cgroup/memory/A/memory.limit_in_bytes > >> 4294967296 > >> > >> $ time cat /export/hdc3/dd_A/tf0 > /dev/zero > >> Killed > >> > > > > I did the same kind of test without any problem...but ok, I'll do more test > > later. > > > > > > > >> real Â0m53.565s > >> user Â0m0.061s > >> sys  0m4.814s > >> > >> Here is the OOM log: > >> > >> May 26 18:43:00 Âkernel: [ Â963.489112] cat invoked oom-killer: > >> gfp_mask=0xd0, order=0, oom_adj=0, oom_score_adj=0 > >> May 26 18:43:00 Âkernel: [ Â963.489121] Pid: 9425, comm: cat Tainted: > >> G    ÂW  2.6.39-mcg-DEV #131 > >> May 26 18:43:00 Âkernel: [ Â963.489123] Call Trace: > >> May 26 18:43:00 Âkernel: [ Â963.489134] Â[<ffffffff810e3512>] > >> dump_header+0x82/0x1af > >> May 26 18:43:00 Âkernel: [ Â963.489137] Â[<ffffffff810e33ca>] ? > >> spin_lock+0xe/0x10 > >> May 26 18:43:00 Âkernel: [ Â963.489140] Â[<ffffffff810e33f9>] ? > >> find_lock_task_mm+0x2d/0x67 > >> May 26 18:43:00 Âkernel: [ Â963.489143] Â[<ffffffff810e38dd>] > >> oom_kill_process+0x50/0x27b > >> May 26 18:43:00 Âkernel: [ Â963.489155] Â[<ffffffff810e3dc6>] > >> mem_cgroup_out_of_memory+0x9a/0xe4 > >> May 26 18:43:00 Âkernel: [ Â963.489160] Â[<ffffffff811153aa>] > >> mem_cgroup_handle_oom+0x134/0x1fe > >> May 26 18:43:00 Âkernel: [ Â963.489163] Â[<ffffffff81114a72>] ? > >> __mem_cgroup_insert_exceeded+0x83/0x83 > >> May 26 18:43:00 Âkernel: [ Â963.489176] Â[<ffffffff811166e9>] > >> __mem_cgroup_try_charge.clone.3+0x368/0x43a > >> May 26 18:43:00 Âkernel: [ Â963.489179] Â[<ffffffff81117586>] > >> mem_cgroup_cache_charge+0x95/0x123 > >> May 26 18:43:00 Âkernel: [ Â963.489183] Â[<ffffffff810e16d8>] > >> add_to_page_cache_locked+0x42/0x114 > >> May 26 18:43:00 Âkernel: [ Â963.489185] Â[<ffffffff810e17db>] > >> add_to_page_cache_lru+0x31/0x5f > >> May 26 18:43:00 Âkernel: [ Â963.489189] Â[<ffffffff81145636>] > >> mpage_readpages+0xb6/0x132 > >> May 26 18:43:00 Âkernel: [ Â963.489194] Â[<ffffffff8119992f>] ? > >> noalloc_get_block_write+0x24/0x24 > >> May 26 18:43:00 Âkernel: [ Â963.489197] Â[<ffffffff8119992f>] ? > >> noalloc_get_block_write+0x24/0x24 > >> May 26 18:43:00 Âkernel: [ Â963.489201] Â[<ffffffff81036742>] ? > >> __switch_to+0x160/0x212 > >> May 26 18:43:00 Âkernel: [ Â963.489205] Â[<ffffffff811978b2>] > >> ext4_readpages+0x1d/0x1f > >> May 26 18:43:00 Âkernel: [ Â963.489209] Â[<ffffffff810e8d4b>] > >> __do_page_cache_readahead+0x144/0x1e3 > >> May 26 18:43:00 Âkernel: [ Â963.489212] Â[<ffffffff810e8e0b>] > >> ra_submit+0x21/0x25 > >> May 26 18:43:00 Âkernel: [ Â963.489215] Â[<ffffffff810e9075>] > >> ondemand_readahead+0x18c/0x19f > >> May 26 18:43:00 Âkernel: [ Â963.489218] Â[<ffffffff810e9105>] > >> page_cache_async_readahead+0x7d/0x86 > >> May 26 18:43:00 Âkernel: [ Â963.489221] Â[<ffffffff810e2b7e>] > >> generic_file_aio_read+0x2d8/0x5fe > >> May 26 18:43:00 Âkernel: [ Â963.489225] Â[<ffffffff81119626>] > >> do_sync_read+0xcb/0x108 > >> May 26 18:43:00 Âkernel: [ Â963.489230] Â[<ffffffff811f168a>] ? > >> fsnotify_perm+0x66/0x72 > >> May 26 18:43:00 Âkernel: [ Â963.489233] Â[<ffffffff811f16f7>] ? > >> security_file_permission+0x2e/0x33 > >> May 26 18:43:00 Âkernel: [ Â963.489236] Â[<ffffffff8111a0c8>] > >> vfs_read+0xab/0x107 > >> May 26 18:43:00 Âkernel: [ Â963.489239] Â[<ffffffff8111a1e4>] sys_read+0x4a/0x6e > >> May 26 18:43:00 Âkernel: [ Â963.489244] Â[<ffffffff8140f469>] > >> sysenter_dispatch+0x7/0x27 > >> May 26 18:43:00 Âkernel: [ Â963.489248] Task in /A killed as a result > >> of limit of /A > >> May 26 18:43:00 Âkernel: [ Â963.489251] memory: usage 4194304kB, limit > >> 4194304kB, failcnt 26 > >> May 26 18:43:00 Âkernel: [ Â963.489253] memory+swap: usage 0kB, limit > >> 9007199254740991kB, failcnt 0 > >> > > > > Hmm, why memory+swap usage 0kb here... > > > > In this set, I used mem_cgroup_margin() rather than res_counter_margin(). > > Hmm, do you disable swap accounting ? If so, I may miss some. > > Yes, I disabled the swap accounting in .config: > # CONFIG_CGROUP_MEM_RES_CTLR_SWAP is not set > > > Here is how i reproduce it: > > $ mkdir /dev/cgroup/memory/D > $ echo 4g >/dev/cgroup/memory/D/memory.limit_in_bytes > > $ cat /dev/cgroup/memory/D/memory.limit_in_bytes > 4294967296 > > $ cat /dev/cgroup/memory/D/memory. > memory.async_control memory.max_usage_in_bytes > memory.soft_limit_in_bytes memory.use_hierarchy > memory.failcnt memory.move_charge_at_immigrate > memory.stat > memory.force_empty memory.oom_control > memory.swappiness > memory.limit_in_bytes memory.reclaim_stat > memory.usage_in_bytes > > $ cat /dev/cgroup/memory/D/memory.async_control > 0 > $ echo 1 >/dev/cgroup/memory/D/memory.async_control > $ cat /dev/cgroup/memory/D/memory.async_control > 1 > > $ echo $$ >/dev/cgroup/memory/D/tasks > $ cat /proc/4358/cgroup > 3:memory:/D > > $ time cat /export/hdc3/dd_A/tf0 > /dev/zero > Killed > If you applied my patches collectly, async_control can be seen if swap controller is configured because of BUG in patch. I could cat 20G file under 4G limit without any problem with boot option swapaccount=0. no problem if async_control == 0 ? Thanks, -Kame -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>