* 2.4.9-ac15 painfully sluggish @ 2001-09-25 15:36 Pau Aliagas 2001-09-25 16:16 ` Rik van Riel 0 siblings, 1 reply; 13+ messages in thread From: Pau Aliagas @ 2001-09-25 15:36 UTC (permalink / raw) To: lkml; +Cc: Alan Cox My impressions of the latest Alan's kernel are very bad, compared to 2.4.9-ac10 which I'm happily running in a Pentium III laptop with 128Mb RAM and 400Mb swap. It hardly touches swap (not at least in top) but the IDE disk never stops. If I run setiathome it's absolutely impossible to do anything at all. Large applications that take seconds to start now take minutes!! Just my 2c Pau ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: 2.4.9-ac15 painfully sluggish 2001-09-25 15:36 2.4.9-ac15 painfully sluggish Pau Aliagas @ 2001-09-25 16:16 ` Rik van Riel 2001-09-25 16:25 ` Pau Aliagas 0 siblings, 1 reply; 13+ messages in thread From: Rik van Riel @ 2001-09-25 16:16 UTC (permalink / raw) To: Pau Aliagas; +Cc: lkml, Alan Cox On Tue, 25 Sep 2001, Pau Aliagas wrote: > My impressions of the latest Alan's kernel are very bad, compared to > 2.4.9-ac10 which I'm happily running in a Pentium III laptop with 128Mb > RAM and 400Mb swap. > > It hardly touches swap (not at least in top) but the IDE disk never stops. > If I run setiathome it's absolutely impossible to do anything at all. > Large applications that take seconds to start now take minutes!! Interesting, the VM changes done to -ac15 seem to have improved performance for all reports I've received up to now. Could you give me some info on how much memory is being used by the various caches (first lines of top) and maybe a few lines of vmstat output ? Lets try to fix this problem... regards, Rik -- IA64: a worthy successor to the i860. http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com/ ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: 2.4.9-ac15 painfully sluggish 2001-09-25 16:16 ` Rik van Riel @ 2001-09-25 16:25 ` Pau Aliagas 2001-09-25 16:37 ` Rik van Riel 2001-09-25 19:32 ` [PATCH] " Rik van Riel 0 siblings, 2 replies; 13+ messages in thread From: Pau Aliagas @ 2001-09-25 16:25 UTC (permalink / raw) To: Rik van Riel; +Cc: lkml, Alan Cox On Tue, 25 Sep 2001, Rik van Riel wrote: > Could you give me some info on how much memory is being > used by the various caches (first lines of top) and maybe > a few lines of vmstat output ? Only setiathome was really running whilst the rest of processes just where in D state; the IDE disk never seemed to stop. Once I stopped the CPU hog (that's seti) kapm-idled ained the CPU, but again the swap was not being used but for a few Kb (about 2M maximum). The problem seems to be related in pages not moved to swap but being discarded somehow and reread later on.... just a guess. If you need any debugging just tell me what and I'll give it a try. Pau ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: 2.4.9-ac15 painfully sluggish 2001-09-25 16:25 ` Pau Aliagas @ 2001-09-25 16:37 ` Rik van Riel 2001-09-25 17:04 ` Pau Aliagas 2001-09-25 19:32 ` [PATCH] " Rik van Riel 1 sibling, 1 reply; 13+ messages in thread From: Rik van Riel @ 2001-09-25 16:37 UTC (permalink / raw) To: Pau Aliagas; +Cc: lkml, Alan Cox On Tue, 25 Sep 2001, Pau Aliagas wrote: > On Tue, 25 Sep 2001, Rik van Riel wrote: > > > Could you give me some info on how much memory is being > > used by the various caches (first lines of top) and maybe > > a few lines of vmstat output ? > If you need any debugging just tell me what and I'll give it a try. Could you send me one screen's worth of output from top and 5 lines from 'vmstat -a 5' ? thanks, Rik -- IA64: a worthy successor to the i860. http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com/ ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: 2.4.9-ac15 painfully sluggish 2001-09-25 16:37 ` Rik van Riel @ 2001-09-25 17:04 ` Pau Aliagas 2001-09-25 17:31 ` Rik van Riel 0 siblings, 1 reply; 13+ messages in thread From: Pau Aliagas @ 2001-09-25 17:04 UTC (permalink / raw) To: Rik van Riel; +Cc: lkml, Alan Cox On Tue, 25 Sep 2001, Rik van Riel wrote: > Could you send me one screen's worth of output from top > and 5 lines from 'vmstat -a 5' ? This is what happens only trying to start evolution for the first time, nothing else. With ac10 works perfectly. With ac15 it takes minutes, at least 4. It was even worst the first time I run the ac15 kernel when updatedb run the find in all the directories. Now I can hardly write this email with pine. [pau@pau pau]$ vmstat -n 5 procs memory swap io system cpu r b w swpd free buff cache si so bi bo in cs us sy id 2 6 1 2392 3040 476 9644 5 7 585 17 170 479 85 6 9 2 12 0 2520 2800 408 9028 25 27 1425 35 208 184 92 6 2 2 4 0 2596 2812 408 9500 15 22 1174 24 204 304 79 6 16 4 5 1 2316 2800 432 9044 47 5 1150 5 227 420 85 7 8 2 4 0 2344 2848 412 8988 30 18 1210 26 238 364 77 11 12 3 3 0 2480 2800 416 8916 15 27 1050 30 188 223 93 6 2 1 3 0 2520 3832 436 8688 22 24 1110 26 195 351 85 5 9 2 9 0 2628 2800 420 9808 4 31 1342 32 191 322 90 8 3 1 10 0 2616 2800 412 10032 33 26 1459 26 241 244 94 6 0 3 7 1 2584 2812 420 10476 10 10 1494 10 206 221 53 6 41 1 7 0 2440 2780 468 9944 49 0 1102 0 187 228 95 5 0 6:58pm up 14 min, 5 users, load average: 7,68, 6,21, 3,22 112 processes: 110 sleeping, 2 running, 0 zombie, 0 stopped CPU states: 2,7% user, 5,2% system, 92,0% nice, 0,0% idle Mem: 127072K av, 124272K used, 2800K free, 564K shrd, 440K buff Swap: 401536K av, 2132K used, 399404K free 9368K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 1008 setiatho 19 19 14796 14M 152 R N 92,4 11,6 11:37 setiathome 4 root 9 0 0 0 0 SW 3,3 0,0 0:20 kswapd 1130 pau 9 0 1916 1916 512 S 1,2 1,5 0:02 sawfish 1516 root 12 0 568 568 324 R 1,2 0,4 0:05 top 1045 root 9 0 25712 16M 616 S 0,9 13,5 0:21 X 1275 pau 9 0 1376 1376 540 D 0,3 1,0 0:13 deskguide_apple 1403 pau 9 0 9512 9512 340 S 0,3 7,4 0:04 galeon-bin 1 root 8 0 100 100 28 S 0,0 0,0 0:04 init 2 root 9 0 0 0 0 SW 0,0 0,0 0:00 keventd 3 root 19 19 0 0 0 SWN 0,0 0,0 0:00 ksoftirqd_CPU0 5 root 9 0 0 0 0 SW 0,0 0,0 0:00 kreclaimd 6 root 9 0 0 0 0 SW 0,0 0,0 0:00 bdflush 7 root 9 0 0 0 0 SW 0,0 0,0 0:00 kupdated 89 root 9 0 0 0 0 SW 0,0 0,0 0:00 khubd 171 root -1 -20 0 0 0 SW< 0,0 0,0 0:00 mdrecoveryd 536 root 9 0 100 100 0 S 0,0 0,0 0:00 syslogd 541 root 9 0 616 616 0 S 0,0 0,4 0:00 klogd 553 rpc 9 0 92 92 0 S 0,0 0,0 0:00 portmap 633 root 9 0 1968 1968 1712 S 0,0 1,5 0:00 ntpd 681 root 9 0 132 132 28 S 0,0 0,1 0:00 automount 693 daemon 9 0 76 76 0 S 0,0 0,0 0:00 atd 708 named 9 0 1756 1756 92 S 0,0 1,3 0:00 named 712 named 9 0 1756 1756 92 S 0,0 1,3 0:00 named 713 named 9 0 1756 1756 92 S 0,0 1,3 0:00 named 714 named 9 0 1756 1756 92 S 0,0 1,3 0:00 named 715 named 9 0 1756 1756 92 S 0,0 1,3 0:00 named 724 root 9 0 200 200 0 S 0,0 0,1 0:00 sshd 738 root 9 0 216 200 0 S 0,0 0,1 0:00 xinetd 760 lp 9 0 152 152 0 S 0,0 0,1 0:00 lpd 775 root 9 0 180 180 0 S 0,0 0,1 0:00 safe_mysqld 803 root 9 0 140 140 56 S 0,0 0,1 0:00 noflushd 821 mysql 9 0 2700 2700 16 S 0,0 2,1 0:00 mysqld 827 mysql 9 0 2700 2700 16 S 0,0 2,1 0:00 mysqld 828 mysql 9 0 2700 2700 16 S 0,0 2,1 0:00 mysqld 836 root 9 0 580 580 40 S 0,0 0,4 0:00 sendmail 847 mysql 9 0 2700 2700 16 S 0,0 2,1 0:00 mysqld 879 junkbust 8 0 524 524 0 S 0,0 0,4 0:00 junkbuster 892 root 9 0 84 84 12 S 0,0 0,0 0:00 gpm 904 root 8 0 152 152 44 S 0,0 0,1 0:00 crond 944 xfs 9 0 3224 3224 0 S 0,0 2,5 0:00 xfs 969 root 9 0 80 80 0 S 0,0 0,0 0:00 rhnsd 1019 root 9 0 0 0 0 SW 0,0 0,0 0:00 kapm-idled 1031 root 9 0 64 64 0 S 0,0 0,0 0:00 mingetty 1032 root 9 0 64 64 0 S 0,0 0,0 0:00 mingetty 1033 root 9 0 64 64 0 S 0,0 0,0 0:00 mingetty 1034 root 9 0 64 64 0 S 0,0 0,0 0:00 mingetty 1035 root 9 0 64 64 0 S 0,0 0,0 0:00 mingetty 1036 root 9 0 64 64 0 S 0,0 0,0 0:00 mingetty 1037 root 9 0 268 268 140 S 0,0 0,2 0:00 gdm Pau ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: 2.4.9-ac15 painfully sluggish 2001-09-25 17:04 ` Pau Aliagas @ 2001-09-25 17:31 ` Rik van Riel 2001-09-25 18:01 ` Pau Aliagas 0 siblings, 1 reply; 13+ messages in thread From: Rik van Riel @ 2001-09-25 17:31 UTC (permalink / raw) To: Pau Aliagas; +Cc: lkml, Alan Cox On Tue, 25 Sep 2001, Pau Aliagas wrote: > On Tue, 25 Sep 2001, Rik van Riel wrote: > > > Could you send me one screen's worth of output from top > > and 5 lines from 'vmstat -a 5' ? > [pau@pau pau]$ vmstat -n 5 > procs memory swap io system cpu > r b w swpd free buff cache si so bi bo in cs us sy id > 2 6 1 2392 3040 476 9644 5 7 585 17 170 479 85 6 9 > 2 12 0 2520 2800 408 9028 25 27 1425 35 208 184 92 6 2 > 2 4 0 2596 2812 408 9500 15 22 1174 24 204 304 79 6 16 Wow, these are a LOT of reads and a very small cache. I guess the old page_launder() might be misbehaving in this case, I've just finished the patch for a new one (which seems to work nicely in the limited tests I've thrown at it). > 6:58pm up 14 min, 5 users, load average: 7,68, 6,21, 3,22 > 112 processes: 110 sleeping, 2 running, 0 zombie, 0 stopped > CPU states: 2,7% user, 5,2% system, 92,0% nice, 0,0% idle > Mem: 127072K av, 124272K used, 2800K free, 564K shrd, 440K buff > Swap: 401536K av, 2132K used, 399404K free 9368K cached Here things get "interesting" ... total memory: 127 MB The memory taken by all your processes: 55 MB buffer + cache: 10 MB ----------------------------------------------------- missing: 62 MB It would be interesting to get a listing of /proc/slabinfo, in particular those lines which have a large number in the 4th or 5th column... regards, Rik -- IA64: a worthy successor to the i860. http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com/ ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: 2.4.9-ac15 painfully sluggish 2001-09-25 17:31 ` Rik van Riel @ 2001-09-25 18:01 ` Pau Aliagas 0 siblings, 0 replies; 13+ messages in thread From: Pau Aliagas @ 2001-09-25 18:01 UTC (permalink / raw) To: Rik van Riel; +Cc: lkml, Alan Cox On Tue, 25 Sep 2001, Rik van Riel wrote: > > [pau@pau pau]$ vmstat -n 5 > > procs memory swap io system cpu > > r b w swpd free buff cache si so bi bo in cs us sy id > > 2 6 1 2392 3040 476 9644 5 7 585 17 170 479 85 6 9 > > 2 12 0 2520 2800 408 9028 25 27 1425 35 208 184 92 6 2 > > 2 4 0 2596 2812 408 9500 15 22 1174 24 204 304 79 6 16 > > Wow, these are a LOT of reads and a very small cache. > Here things get "interesting" ... > > total memory: 127 MB > The memory taken by all your processes: 55 MB > buffer + cache: 10 MB > ----------------------------------------------------- > missing: 62 MB > > It would be interesting to get a listing of /proc/slabinfo, > in particular those lines which have a large number in the > 4th or 5th column... 1. Before starting evolution: slabinfo - version: 1.1 kmem_cache 59 68 112 2 2 1 ip_conntrack 11 33 352 3 3 1 ip_fib_hash 12 113 32 1 1 1 uhci_urb_priv 1 67 56 1 1 1 ip_mrt_cache 0 0 96 0 0 1 tcp_tw_bucket 1 40 96 1 1 1 tcp_bind_bucket 12 113 32 1 1 1 tcp_open_request 0 0 64 0 0 1 inet_peer_cache 1 59 64 1 1 1 ip_dst_cache 41 100 192 5 5 1 arp_cache 2 30 128 1 1 1 blkdev_requests 512 520 96 13 13 1 dnotify cache 0 0 20 0 0 1 file lock cache 4 42 92 1 1 1 fasync cache 1 202 16 1 1 1 uid_cache 7 113 32 1 1 1 skbuff_head_cache 129 240 160 10 10 1 sock 300 333 832 37 37 2 sigqueue 1024 1044 132 36 36 1 kiobuf 0 0 8768 0 0 4 cdev_cache 23 118 64 2 2 1 bdev_cache 6 59 64 1 1 1 mnt_cache 16 59 64 1 1 1 inode_cache 1036 2178 416 242 242 1 dentry_cache 1019 2730 128 91 91 1 dquot 0 0 128 0 0 1 filp 3116 3120 96 78 78 1 names_cache 0 1 4096 0 1 1 buffer_head 3816 5240 96 131 131 1 mm_struct 83 96 160 4 4 1 vm_area_struct 4536 4897 64 83 83 1 fs_cache 82 118 64 2 2 1 files_cache 82 90 416 10 10 1 signal_act 87 90 1312 30 30 1 size-131072(DMA) 0 0 131072 0 0 32 size-131072 0 0 131072 0 0 32 size-65536(DMA) 0 0 65536 0 0 16 size-65536 0 0 65536 0 0 16 size-32768(DMA) 0 0 32768 0 0 8 size-32768 1 1 32768 1 1 8 size-16384(DMA) 0 0 16384 0 0 4 size-16384 0 0 16384 0 0 4 size-8192(DMA) 0 0 8192 0 0 2 size-8192 6 6 8192 6 6 2 size-4096(DMA) 0 0 4096 0 0 1 size-4096 33 33 4096 33 33 1 size-2048(DMA) 0 0 2048 0 0 1 size-2048 7 8 2048 4 4 1 size-1024(DMA) 0 0 1024 0 0 1 size-1024 57 60 1024 15 15 1 size-512(DMA) 0 0 512 0 0 1 size-512 16 24 512 2 3 1 size-256(DMA) 0 0 256 0 0 1 size-256 21 30 256 2 2 1 size-128(DMA) 2 30 128 1 1 1 size-128 462 510 128 17 17 1 size-64(DMA) 0 0 64 0 0 1 size-64 242 295 64 5 5 1 size-32(DMA) 2 113 32 1 1 1 size-32 429 1017 32 9 9 1 evol 2.while starting evolution (evolution-pine-importer looks through MANY pine mboxes to see if it can import them, more or less like updatedb that ckecks all the files in the disc, so it looks like something related to reading lots of files) slabinfo - version: 1.1 kmem_cache 59 68 112 2 2 1 ip_conntrack 7 33 352 3 3 1 ip_fib_hash 12 113 32 1 1 1 uhci_urb_priv 1 67 56 1 1 1 ip_mrt_cache 0 0 96 0 0 1 tcp_tw_bucket 1 40 96 1 1 1 tcp_bind_bucket 12 113 32 1 1 1 tcp_open_request 0 0 64 0 0 1 inet_peer_cache 1 59 64 1 1 1 ip_dst_cache 55 100 192 5 5 1 arp_cache 2 30 128 1 1 1 blkdev_requests 512 520 96 13 13 1 dnotify cache 0 0 20 0 0 1 file lock cache 4 42 92 1 1 1 fasync cache 1 202 16 1 1 1 uid_cache 7 113 32 1 1 1 skbuff_head_cache 161 240 160 10 10 1 sock 324 324 832 36 36 2 sigqueue 1024 1044 132 36 36 1 kiobuf 0 0 8768 0 0 4 cdev_cache 21 118 64 2 2 1 bdev_cache 6 59 64 1 1 1 mnt_cache 16 59 64 1 1 1 inode_cache 994 2178 416 242 242 1 dentry_cache 984 2730 128 91 91 1 dquot 0 0 128 0 0 1 filp 3116 3120 96 78 78 1 names_cache 1 4 4096 1 4 1 buffer_head 2229 3440 96 86 86 1 mm_struct 86 96 160 4 4 1 vm_area_struct 4929 4956 64 84 84 1 fs_cache 85 118 64 2 2 1 files_cache 85 90 416 10 10 1 signal_act 90 93 1312 31 31 1 size-131072(DMA) 0 0 131072 0 0 32 size-131072 0 0 131072 0 0 32 size-65536(DMA) 0 0 65536 0 0 16 size-65536 0 0 65536 0 0 16 size-32768(DMA) 0 0 32768 0 0 8 size-32768 1 2 32768 1 2 8 size-16384(DMA) 0 0 16384 0 0 4 size-16384 0 0 16384 0 0 4 size-8192(DMA) 0 0 8192 0 0 2 size-8192 6 6 8192 6 6 2 size-4096(DMA) 0 0 4096 0 0 1 size-4096 37 38 4096 37 38 1 size-2048(DMA) 0 0 2048 0 0 1 size-2048 9 12 2048 5 6 1 size-1024(DMA) 0 0 1024 0 0 1 size-1024 57 60 1024 15 15 1 size-512(DMA) 0 0 512 0 0 1 size-512 20 32 512 4 4 1 size-256(DMA) 0 0 256 0 0 1 size-256 39 60 256 4 4 1 size-128(DMA) 2 30 128 1 1 1 size-128 503 570 128 19 19 1 size-64(DMA) 0 0 64 0 0 1 size-64 248 295 64 5 5 1 size-32(DMA) 2 113 32 1 1 1 size-32 442 1017 32 9 9 1 BTW, It has taken me more than 10 minutes to write this email and evolution hasn't yet started. 3.last slabinfo (the disk still "plays") slabinfo - version: 1.1 kmem_cache 59 68 112 2 2 1 ip_conntrack 10 22 352 2 2 1 ip_fib_hash 12 113 32 1 1 1 uhci_urb_priv 1 67 56 1 1 1 ip_mrt_cache 0 0 96 0 0 1 tcp_tw_bucket 1 40 96 1 1 1 tcp_bind_bucket 11 113 32 1 1 1 tcp_open_request 0 0 64 0 0 1 inet_peer_cache 2 59 64 1 1 1 ip_dst_cache 45 100 192 5 5 1 arp_cache 2 30 128 1 1 1 blkdev_requests 512 520 96 13 13 1 dnotify cache 0 0 20 0 0 1 file lock cache 9 42 92 1 1 1 fasync cache 1 202 16 1 1 1 uid_cache 7 113 32 1 1 1 skbuff_head_cache 129 264 160 11 11 1 sock 344 351 832 39 39 2 sigqueue 433 435 132 15 15 1 kiobuf 0 0 8768 0 0 4 cdev_cache 21 118 64 2 2 1 bdev_cache 6 59 64 1 1 1 mnt_cache 16 59 64 1 1 1 inode_cache 1023 2178 416 242 242 1 dentry_cache 1012 2730 128 91 91 1 dquot 0 0 128 0 0 1 filp 3116 3120 96 78 78 1 names_cache 0 3 4096 0 3 1 buffer_head 2072 2520 96 63 63 1 mm_struct 84 96 160 4 4 1 vm_area_struct 5156 5251 64 89 89 1 fs_cache 83 118 64 2 2 1 files_cache 83 90 416 10 10 1 signal_act 88 93 1312 31 31 1 size-131072(DMA) 0 0 131072 0 0 32 size-131072 0 0 131072 0 0 32 size-65536(DMA) 0 0 65536 0 0 16 size-65536 0 0 65536 0 0 16 size-32768(DMA) 0 0 32768 0 0 8 size-32768 1 1 32768 1 1 8 size-16384(DMA) 0 0 16384 0 0 4 size-16384 0 0 16384 0 0 4 size-8192(DMA) 0 0 8192 0 0 2 size-8192 6 6 8192 6 6 2 size-4096(DMA) 0 0 4096 0 0 1 size-4096 33 34 4096 33 34 1 size-2048(DMA) 0 0 2048 0 0 1 size-2048 7 10 2048 4 5 1 size-1024(DMA) 0 0 1024 0 0 1 size-1024 53 56 1024 14 14 1 size-512(DMA) 0 0 512 0 0 1 size-512 16 32 512 2 4 1 size-256(DMA) 0 0 256 0 0 1 size-256 21 30 256 2 2 1 size-128(DMA) 2 30 128 1 1 1 size-128 461 510 128 17 17 1 size-64(DMA) 0 0 64 0 0 1 size-64 249 295 64 5 5 1 size-32(DMA) 2 113 32 1 1 1 size-32 440 1017 32 9 9 1 While the disc is reading: procs memory swap io system cpu r b w swpd free buff cache si so bi bo in cs us sy id 3 11 0 2492 2800 436 8300 9 8 610 14 160 308 89 6 6 2 7 0 2412 3568 408 8232 20 2 1538 4 222 264 76 9 15 1 10 0 2664 2800 432 9128 3 30 1388 40 212 201 89 8 3 2 7 0 2616 3020 492 9360 26 25 1457 25 204 198 84 6 10 4 10 0 2648 2800 420 8780 8 0 1377 6 201 233 88 7 4 1 10 0 2528 2800 440 8736 9 0 1248 0 220 280 82 6 12 1 6 1 2520 2800 448 8688 1 3 1350 30 212 230 75 12 13 Pau ^ permalink raw reply [flat|nested] 13+ messages in thread
* [PATCH] Re: 2.4.9-ac15 painfully sluggish 2001-09-25 16:25 ` Pau Aliagas 2001-09-25 16:37 ` Rik van Riel @ 2001-09-25 19:32 ` Rik van Riel 2001-09-25 20:42 ` Jasper Spaans ` (3 more replies) 1 sibling, 4 replies; 13+ messages in thread From: Rik van Riel @ 2001-09-25 19:32 UTC (permalink / raw) To: Pau Aliagas; +Cc: lkml, Alan Cox, Roberto Orenstein, Francois Romieu On Tue, 25 Sep 2001, Pau Aliagas wrote: > The problem seems to be related in pages not moved to swap but > being discarded somehow and reread later on.... just a guess. I've made a small patch to 2.4.9-ac15 which should make page_launder() smoother, make some (very minor) tweaks to page aging and updates various comments in vmscan.c It's below this email and at: http://www.surriel.com/patches/2.4/2.4.9-ac15-age+launder Since I failed to break 2.4.9-ac15 with this patch following the instructions given to me by others, it would be nice to know if the thing can still break on your machines. Please test, Rik -- IA64: a worthy successor to the i860. http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com/ --- linux-2.4.9-ac15/mm/vmscan.c.orig Mon Sep 24 17:30:48 2001 +++ linux-2.4.9-ac15/mm/vmscan.c Tue Sep 25 13:33:16 2001 @@ -28,19 +28,12 @@ static inline void age_page_up(struct page *page) { - unsigned long age = page->age; - age += PAGE_AGE_ADV; - if (age > PAGE_AGE_MAX) - age = PAGE_AGE_MAX; - page->age = age; + page->age = min((int) (page->age + PAGE_AGE_ADV), PAGE_AGE_MAX); } static inline void age_page_down(struct page *page) { - unsigned long age = page->age; - if (age > 0) - age -= PAGE_AGE_DECL; - page->age = age; + page->age -= min(PAGE_AGE_DECL, (int)page->age); } /* @@ -108,6 +101,23 @@ pte_t pte; swp_entry_t entry; + /* Don't look at this pte if it's been accessed recently. */ + if (ptep_test_and_clear_young(page_table)) { + age_page_up(page); + return; + } + + /* + * If the page is on the active list, page aging is done in + * refill_inactive_scan(), anonymous pages are aged here. + * This is done so heavily shared pages (think libc.so) + * don't get punished heavily while they are still in use. + * The alternative would be to put anonymous pages on the + * active list too, but that increases complexity (for now). + */ + if (!PageActive(page)) + age_page_down(page); + /* * If we have plenty inactive pages on this * zone, skip it. @@ -133,23 +143,6 @@ pte = ptep_get_and_clear(page_table); flush_tlb_page(vma, address); - /* Don't look at this pte if it's been accessed recently. */ - if (ptep_test_and_clear_young(page_table)) { - age_page_up(page); - return; - } - - /* - * If the page is on the active list, page aging is done in - * refill_inactive_scan(), anonymous pages are aged here. - * This is done so heavily shared pages (think libc.so) - * don't get punished heavily while they are still in use. - * The alternative would be to put anonymous pages on the - * active list too, but that increases complexity (for now). - */ - if (!PageActive(page)) - age_page_down(page); - /* * Is the page already in the swap cache? If so, then * we can just drop our reference to it without doing @@ -447,6 +440,7 @@ page_count(page) > (1 + !!page->buffers)) { del_page_from_inactive_clean_list(page); add_page_to_active_list(page); + page->age = max((int)page->age, PAGE_AGE_START); continue; } @@ -536,34 +530,56 @@ * @gfp_mask: what operations we are allowed to do * @sync: are we allowed to do synchronous IO in emergencies ? * - * When this function is called, we are most likely low on free + - * inactive_clean pages. Since we want to refill those pages as - * soon as possible, we'll make two loops over the inactive list, - * one to move the already cleaned pages to the inactive_clean lists - * and one to (often asynchronously) clean the dirty inactive pages. + * This function is called when we are low on free / inactive_clean + * pages, its purpose is to refill the free/clean list as efficiently + * as possible. * - * In situations where kswapd cannot keep up, user processes will - * end up calling this function. Since the user process needs to - * have a page before it can continue with its allocation, we'll - * do synchronous page flushing in that case. + * This means we do writes asynchronously as long as possible and will + * only sleep on IO when we don't have another option. Since writeouts + * cause disk seeks and make read IO slower, we skip writes alltogether + * when the amount of dirty pages is small. * * This code is heavily inspired by the FreeBSD source code. Thanks * go out to Matthew Dillon. */ -#define MAX_LAUNDER (4 * (1 << page_cluster)) -#define CAN_DO_FS (gfp_mask & __GFP_FS) -#define CAN_DO_IO (gfp_mask & __GFP_IO) +#define CAN_DO_FS ((gfp_mask & __GFP_FS) && should_write) +#define WRITE_LOW_WATER 5 +#define WRITE_HIGH_WATER 10 int page_launder(int gfp_mask, int sync) { - int launder_loop, maxscan, cleaned_pages, maxlaunder; + int maxscan, cleaned_pages, failed_pages, total; struct list_head * page_lru; struct page * page; - launder_loop = 0; - maxlaunder = 0; + /* + * This flag determines if we should do writeouts of dirty data + * or not. When more than WRITE_HIGH_WATER percentage of the + * pages we process would need to be written out we set this flag + * and will do writeout, the flag is cleared once we go below + * WRITE_LOW_WATER. Note that only pages we actually process + * get counted, ie. pages where we make it beyond the TryLock. + * + * XXX: These flags still need tuning. + */ + static int should_write = 0; + cleaned_pages = 0; + failed_pages = 0; + + /* + * The gfp_mask tells try_to_free_buffers() below if it should + * wait do IO or may be allowed to wait on IO synchronously. + * + * Note that syncronous IO only happens when a page has not been + * written out yet when we see it for a second time, this is done + * through magic in try_to_free_buffers(). + */ + if (!should_write) + gfp_mask &= ~(__GFP_WAIT | __GFP_IO); + else if (!sync) + gfp_mask &= ~__GFP_WAIT; -dirty_page_rescan: + /* The main launder loop. */ spin_lock(&pagemap_lru_lock); maxscan = nr_inactive_dirty_pages; while ((page_lru = inactive_dirty_list.prev) != &inactive_dirty_list && @@ -580,26 +596,36 @@ } /* - * If the page is or was in use, we move it to the active - * list. Note that pages in use by processes can easily - * end up here to be unmapped later, but we just don't - * want them clogging up the inactive list. + * The page is in active use or really unfreeable. Move to + * the active list and adjust the page age if needed. */ - if ((PageReferenced(page) || page->age > 0 || - page_count(page) > (1 + !!page->buffers) || - page_ramdisk(page)) && !PageLocked(page)) { + if (PageReferenced(page) || page->age || page_ramdisk(page)) { del_page_from_inactive_dirty_list(page); add_page_to_active_list(page); + page->age = max((int)page->age, PAGE_AGE_START); continue; } /* - * If we have plenty free pages on a zone: + * The page is still in the page tables of some process, + * move it to the active list but leave page age at 0; + * either swap_out() will make it freeable soon or it is + * mlock()ed... * - * 1) we avoid a writeout for that page if its dirty. - * 2) if its a buffercache page, and not a pagecache - * one, we skip it since we cannot move it to the - * inactive clean list --- we have to free it. + * The !PageLocked() test is to protect us from ourselves, + * see the code around the writepage() call. + */ + if ((page_count(page) > (1 + !!page->buffers)) && + !PageLocked(page)) { + del_page_from_inactive_dirty_list(page); + add_page_to_active_list(page); + continue; + } + + /* + * If this zone has plenty of pages free, don't spend time + * on cleaning it but only move clean pages out of the way + * so we won't have to scan those again. */ if (zone_free_plenty(page->zone)) { if (!page->mapping || page_dirty(page)) { @@ -633,11 +659,12 @@ if (!writepage) goto page_active; - /* First time through? Move it to the back of the list */ - if (!launder_loop || !CAN_DO_FS) { + /* Can't do it? Move it to the back of the list. */ + if (!CAN_DO_FS) { list_del(page_lru); list_add(page_lru, &inactive_dirty_list); UnlockPage(page); + failed_pages++; continue; } @@ -664,9 +691,7 @@ * buffer pages */ if (page->buffers) { - unsigned int buffer_mask; int clearedbuf; - int freed_page = 0; /* * Since we might be doing disk IO, we have to * drop the spinlock and take an extra reference @@ -676,16 +701,8 @@ page_cache_get(page); spin_unlock(&pagemap_lru_lock); - /* Will we do (asynchronous) IO? */ - if (launder_loop && maxlaunder == 0 && sync) - buffer_mask = gfp_mask; /* Do as much as we can */ - else if (launder_loop && maxlaunder-- > 0) - buffer_mask = gfp_mask & ~__GFP_WAIT; /* Don't wait, async write-out */ - else - buffer_mask = gfp_mask & ~(__GFP_WAIT | __GFP_IO); /* Don't even start IO */ - - /* Try to free metadata underlying the page. */ - clearedbuf = try_to_release_page(page, buffer_mask); + /* Try to free the page buffers. */ + clearedbuf = try_to_release_page(page, gfp_mask); /* * Re-take the spinlock. Note that we cannot @@ -697,11 +714,11 @@ /* The buffers were not freed. */ if (!clearedbuf) { add_page_to_inactive_dirty_list(page); + failed_pages++; /* The page was only in the buffer cache. */ } else if (!page->mapping) { atomic_dec(&buffermem_pages); - freed_page = 1; cleaned_pages++; /* The page has more users besides the cache and us. */ @@ -722,13 +739,6 @@ UnlockPage(page); page_cache_release(page); - /* - * If we're freeing buffer cache pages, stop when - * we've got enough free memory. - */ - if (freed_page && !free_shortage()) - break; - continue; } else if (page->mapping && !PageDirty(page)) { /* @@ -750,33 +760,22 @@ */ del_page_from_inactive_dirty_list(page); add_page_to_active_list(page); + page->age = max((int)page->age, PAGE_AGE_START); UnlockPage(page); } } spin_unlock(&pagemap_lru_lock); /* - * If we don't have enough free pages, we loop back once - * to queue the dirty pages for writeout. When we were called - * by a user process (that /needs/ a free page) and we didn't - * free anything yet, we wait synchronously on the writeout of - * MAX_SYNC_LAUNDER pages. - * - * We also wake up bdflush, since bdflush should, under most - * loads, flush out the dirty pages before we have to wait on - * IO. + * Set the should_write flag, for the next callers of page_launder. + * If we go below the low watermark we stop the writeout of dirty + * pages, writeout is started when we get above the high watermark. */ - if (CAN_DO_IO && !launder_loop && free_shortage()) { - launder_loop = 1; - /* If we cleaned pages, never do synchronous IO. */ - if (cleaned_pages) - sync = 0; - /* We only do a few "out of order" flushes. */ - maxlaunder = MAX_LAUNDER; - /* Kflushd takes care of the rest. */ - wakeup_bdflush(0); - goto dirty_page_rescan; - } + total = failed_pages + cleaned_pages; + if (should_write && failed_pages * 100 < WRITE_LOW_WATER * total) + should_write = 0; + else if (!should_write && failed_pages * 100 > WRITE_HIGH_WATER * total) + should_write = 1; /* Return the number of pages moved to the inactive_clean list. */ return cleaned_pages; @@ -998,27 +997,26 @@ return progress; } - - +/* + * Worker function for kswapd and try_to_free_pages, we get + * called whenever there is a shortage of free/inactive_clean + * pages. + * + * This function will also move pages to the inactive list, + * if needed. + */ static int do_try_to_free_pages(unsigned int gfp_mask, int user) { int ret = 0; /* - * If we're low on free pages, move pages from the - * inactive_dirty list to the inactive_clean list. - * - * Usually bdflush will have pre-cleaned the pages - * before we get around to moving them to the other - * list, so this is a relatively cheap operation. + * Eat memory from filesystem page cache, buffer cache, + * dentry, inode and filesystem quota caches. */ - - if (free_shortage()) { - ret += page_launder(gfp_mask, user); - shrink_dcache_memory(0, gfp_mask); - shrink_icache_memory(0, gfp_mask); - shrink_dqcache_memory(DEF_PRIORITY, gfp_mask); - } + ret += page_launder(gfp_mask, user); + shrink_dcache_memory(0, gfp_mask); + shrink_icache_memory(0, gfp_mask); + shrink_dqcache_memory(DEF_PRIORITY, gfp_mask); /* * If needed, we move pages from the active list @@ -1028,7 +1026,7 @@ ret += refill_inactive(gfp_mask); /* - * Reclaim unused slab cache if memory is low. + * Reclaim unused slab cache memory. */ kmem_cache_reap(gfp_mask); @@ -1080,7 +1078,7 @@ static long recalc = 0; /* If needed, try to free some memory. */ - if (inactive_shortage() || free_shortage()) + if (free_shortage()) do_try_to_free_pages(GFP_KSWAPD, 0); /* Once a second ... */ @@ -1090,8 +1088,6 @@ /* Do background page aging. */ refill_inactive_scan(DEF_PRIORITY); } - - run_task_queue(&tq_disk); /* * We go to sleep if either the free page shortage ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Re: 2.4.9-ac15 painfully sluggish 2001-09-25 19:32 ` [PATCH] " Rik van Riel @ 2001-09-25 20:42 ` Jasper Spaans 2001-09-25 20:52 ` Rik van Riel 2001-09-26 7:36 ` Pau Aliagas ` (2 subsequent siblings) 3 siblings, 1 reply; 13+ messages in thread From: Jasper Spaans @ 2001-09-25 20:42 UTC (permalink / raw) To: Rik van Riel Cc: Pau Aliagas, lkml, Alan Cox, Roberto Orenstein, Francois Romieu On Tue, Sep 25, 2001 at 04:32:13PM -0300, Rik van Riel wrote: > > The problem seems to be related in pages not moved to swap but > > being discarded somehow and reread later on.... just a guess. > > I've made a small patch to 2.4.9-ac15 which should make > page_launder() smoother, make some (very minor) tweaks > to page aging and updates various comments in vmscan.c > > It's below this email and at: [snip] With this -painfully applied- patch, all seems a lot better. However, it still seems more sluggish than 2.4.9-ac12. Haven't bothered testing -ac13 and -ac14. Swapping still seems to be a problem though, as this kernel happily claims several tens of MBs of swap, and then, out of the blue, starts writing pages out in huge blocks. The vmstat info just disappeared from my screen, while stressing the vm-system by doing a large parallel build, however, it showed no paging happening during long times (20-30 secs) and then suddenly writing 30 MB in one go. Doesn't seem very smooth to me, where can I get my refund? :) (if you'd like more details, please ask, I can do some testing) Regards, -- Q_. Jasper Spaans <jasper@spaans.ds9a.nl> `~\ http://jsp.ds9a.nl/ Mr /\ Tel/Fax: +31-84-8749842 Zap Move '.sig' for great justice! ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Re: 2.4.9-ac15 painfully sluggish 2001-09-25 20:42 ` Jasper Spaans @ 2001-09-25 20:52 ` Rik van Riel 0 siblings, 0 replies; 13+ messages in thread From: Rik van Riel @ 2001-09-25 20:52 UTC (permalink / raw) To: Jasper Spaans Cc: Pau Aliagas, lkml, Alan Cox, Roberto Orenstein, Francois Romieu On Tue, 25 Sep 2001, Jasper Spaans wrote: > (if you'd like more details, please ask, I can do some testing) Some 'vmstat -a' output would be helpful, as well as a screen of top. Rik -- IA64: a worthy successor to the i860. http://www.surriel.com/ http://www.conectiva.com/ http://distro.conectiva.com/ ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Re: 2.4.9-ac15 painfully sluggish 2001-09-25 19:32 ` [PATCH] " Rik van Riel 2001-09-25 20:42 ` Jasper Spaans @ 2001-09-26 7:36 ` Pau Aliagas 2001-09-26 13:54 ` [PATCH] " Roberto Orenstein 2001-09-26 18:01 ` [PATCH] " Kent Borg 3 siblings, 0 replies; 13+ messages in thread From: Pau Aliagas @ 2001-09-26 7:36 UTC (permalink / raw) To: Rik van Riel; +Cc: lkml, Alan Cox, Roberto Orenstein, Francois Romieu On Tue, 25 Sep 2001, Rik van Riel wrote: > On Tue, 25 Sep 2001, Pau Aliagas wrote: > > I've made a small patch to 2.4.9-ac15 which should make > page_launder() smoother, make some (very minor) tweaks > to page aging and updates various comments in vmscan.c Ok, things improve vastly and the system is usable again. I cannot "notice" any difference with ac10 in terms of speed. I attach the top and vmstat outputs while running updatedb (reads all the files in the filesystems) and starting evolution for the first time (touches all the mail folders, that's hundreths of files). It starts smoothly in seconds. 9:30am up 52 min, 5 users, load average: 2,83, 2,35, 1,87 125 processes: 121 sleeping, 4 running, 0 zombie, 0 stopped CPU states: 16,2% user, 7,9% system, 75,8% nice, 0,0% idle Mem: 127072K av, 123336K used, 3736K free, 2360K shrd, 7988K buff Swap: 401536K av, 54092K used, 347444K free 33952K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 1008 setiatho 20 19 15348 14M 708 R N 76,2 11,4 38:28 setiathome 1045 root 9 0 29740 12M 4076 S 7,3 9,6 9:05 X 2238 pau 9 0 8876 8876 6764 D 4,1 6,9 0:00 evolution-mail 1278 pau 9 0 4276 3776 3444 S 2,5 2,9 0:45 deskguide_apple 2192 pau 9 0 16272 15M 6064 S 1,9 12,8 0:01 evolution 1176 pau 9 0 1784 1456 1352 R 1,5 1,1 0:01 xscreensaver 4 root 9 0 0 0 0 SW 0,9 0,0 0:02 kswapd 2189 pau 13 0 1096 1096 840 R 0,9 0,8 0:00 top 2258 pau 9 0 6024 6024 4632 S 0,9 4,7 0:00 evolution-pine- 2198 pau 9 0 5616 5616 4360 D 0,7 4,4 0:00 wombat 1127 pau 9 0 3424 3208 2032 S 0,3 2,5 0:03 sawfish 1188 pau 9 0 5428 4972 4172 S 0,3 3,9 0:02 panel 2265 pau 9 0 5372 5372 4216 S 0,3 4,2 0:00 evolution-gnome 944 xfs 9 0 4748 2160 1768 D 0,1 1,6 0:01 xfs 1198 pau 9 0 5232 3992 3472 S 0,1 3,1 0:02 gnome-terminal 1944 root 19 19 716 676 496 R N 0,1 0,5 0:01 updatedb 2093 pau 9 0 524 496 444 S 0,1 0,3 0:00 vmstat procs memory swap io system cpu r b w swpd free buff cache si so bi bo in cs us sy id 4 1 0 57552 3620 14280 42892 9 18 78 28 124 881 95 3 2 1 11 0 54756 8220 14452 40896 386 115 614 163 244 560 96 4 0 3 0 0 54660 7208 14720 41432 234 0 394 0 180 358 97 3 0 5 0 0 54660 4952 15196 42508 74 0 381 0 162 295 98 2 0 3 2 0 53856 5680 13244 42476 197 2 428 14 205 443 95 5 0 2 4 0 53856 3668 13668 43212 170 0 395 6 180 399 97 3 0 1 3 0 54004 5852 10324 42724 114 0 368 1 176 841 93 6 1 4 0 0 54000 2800 10640 41908 19 0 314 2 151 764 95 5 0 3 0 0 54024 6788 7568 39048 1 0 350 6 172 497 95 5 0 1 2 0 54024 4656 7780 40192 78 0 347 11 163 524 96 4 0 4 0 0 53972 2936 7944 38844 230 0 420 4 193 1370 94 6 0 3 0 0 53972 12944 8420 39136 63 0 285 11 198 1262 96 4 0 4 0 0 53972 11816 8836 39644 19 0 242 14 164 426 97 3 0 4 0 0 53972 10532 9216 40252 31 0 222 17 215 834 96 4 0 3 1 0 53972 8628 9504 40944 42 0 235 4 212 514 96 4 0 2 1 0 53972 4040 9732 41636 0 0 173 2 210 798 95 5 0 4 2 0 54160 3096 7188 40800 32 0 207 32 134 1502 87 13 0 3 0 0 54332 3732 6412 35452 10 0 242 16 178 773 96 4 0 4 0 0 54332 2800 7060 33908 0 0 232 16 204 506 96 4 0 4 0 0 54072 4616 7096 34748 0 572 211 679 207 436 97 3 0 3 2 0 54024 2800 7280 35364 0 12 244 18 216 726 97 3 0 7 0 0 54020 3104 7644 34116 41 0 229 0 175 954 93 7 0 3 1 0 54000 3744 7888 33852 62 0 374 0 193 722 97 3 0 2 4 0 54092 4952 7988 33856 8 0 118 180 170 2934 91 9 0 4 1 0 54036 2992 8024 34312 38 0 102 444 237 1684 92 8 0 2 0 0 54036 3580 8216 35092 0 0 208 0 173 356 97 3 0 3 0 0 55016 3820 8444 34912 3 67 293 136 165 279 90 2 8 3 0 0 55016 2800 8724 34820 0 0 202 15 177 385 98 2 0 4 0 0 55016 2948 9008 34704 0 0 258 14 175 372 98 2 0 Pau ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] 2.4.9-ac15 painfully sluggish 2001-09-25 19:32 ` [PATCH] " Rik van Riel 2001-09-25 20:42 ` Jasper Spaans 2001-09-26 7:36 ` Pau Aliagas @ 2001-09-26 13:54 ` Roberto Orenstein 2001-09-26 18:01 ` [PATCH] " Kent Borg 3 siblings, 0 replies; 13+ messages in thread From: Roberto Orenstein @ 2001-09-26 13:54 UTC (permalink / raw) To: linux-kernel Rik van Riel wrote: > On Tue, 25 Sep 2001, Pau Aliagas wrote: > > >> The problem seems to be related in pages not moved to swap but >> being discarded somehow and reread later on.... just a guess. > > > I've made a small patch to 2.4.9-ac15 which should make > page_launder() smoother, make some (very minor) tweaks > to page aging and updates various comments in vmscan.c > > It's below this email and at: > > http://www.surriel.com/patches/2.4/2.4.9-ac15-age+launder > > Since I failed to break 2.4.9-ac15 with this patch following > the instructions given to me by others, it would be nice to > know if the thing can still break on your machines. > > Please test, > Test done, and it seems just fine. The problem vanish away. Didn't trigger anymore, on my machine. thanx Roberto ^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [PATCH] Re: 2.4.9-ac15 painfully sluggish 2001-09-25 19:32 ` [PATCH] " Rik van Riel ` (2 preceding siblings ...) 2001-09-26 13:54 ` [PATCH] " Roberto Orenstein @ 2001-09-26 18:01 ` Kent Borg 3 siblings, 0 replies; 13+ messages in thread From: Kent Borg @ 2001-09-26 18:01 UTC (permalink / raw) To: Rik van Riel; +Cc: lkml On Tue, Sep 25, 2001 at 04:32:13PM -0300, Rik van Riel wrote: > I've made a small patch to 2.4.9-ac15 which should make > page_launder() smoother, make some (very minor) tweaks > to page aging and updates various comments in vmscan.c Works for me. I have the premption patch turned on. On a 192 MB PIII laptop running at 500 MHz I have two X sessions running, and have opened a zillion application windows on each, including over a dozen Netscape windows, some Mozilla's, a couple emacs's, a kernel compile, Staroffice. About every app I can immediately think of. I haven't tried malloc-ing a ton of memory, but this contrived real-world test works. Swap is at 238 MB. Lower than I would have expected, but that doesn't mean I know anything. Switching between X sessions just took near 10-seconds going, and only about 5-seconds coming back. Switching in Staroffice is painful, but generally the responsiveness feels quite nice. I like it. Thanks, -kb, the Kent who will be staying with 2.4.9-ac15, plus preemption patch, plus this patch--once he figures out how to close all those windows. ^ permalink raw reply [flat|nested] 13+ messages in thread
end of thread, other threads:[~2001-09-26 18:01 UTC | newest] Thread overview: 13+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2001-09-25 15:36 2.4.9-ac15 painfully sluggish Pau Aliagas 2001-09-25 16:16 ` Rik van Riel 2001-09-25 16:25 ` Pau Aliagas 2001-09-25 16:37 ` Rik van Riel 2001-09-25 17:04 ` Pau Aliagas 2001-09-25 17:31 ` Rik van Riel 2001-09-25 18:01 ` Pau Aliagas 2001-09-25 19:32 ` [PATCH] " Rik van Riel 2001-09-25 20:42 ` Jasper Spaans 2001-09-25 20:52 ` Rik van Riel 2001-09-26 7:36 ` Pau Aliagas 2001-09-26 13:54 ` [PATCH] " Roberto Orenstein 2001-09-26 18:01 ` [PATCH] " Kent Borg
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox