Mem-info:
Free pages: 548kB
( Free: 137 (48 96 144)
15*4kB 9*8kB 6*16kB 2*32kB 0*64kB 2*128kB = 548kB)
Swap cache: add 14096/14096, delete 14032/14032, find 0/0
Free swap: 28612kB
1536 pages of RAM
345 reserved pages
231 pages shared
64 pages swap cached
23 pages in page table cache
Buffer memory: 184kB
Buffer heads: 245
Buffer blocks: 184
CLEAN: 144 buffers, 1 used (last=2), 0 locked, 0 protected, 0 dirty
LOCKED: 33 buffers, 20 used (last=31), 0 locked, 0 protected, 0 dirty
DIRTY: 3 buffers, 0 used (last=0), 0 locked, 0 protected, 3 dirty
Networking buffers in use : 17
Total network buffer allocations : 505
Total failed network buffer allocs : 0
IP fragment buffer size : 0
Did:
echo "0 0" >/proc/sys/vm/pagetable_cache
Free pages: 576kB
( Free: 144 (48 96 144)
28*4kB 4*8kB 5*16kB 3*32kB 0*64kB 2*128kB = 576kB)
Swap cache: add 21266/21266, delete 21238/21238, find 0/0
Free swap: 28296kB
1536 pages of RAM
345 reserved pages
239 pages shared
28 pages swap cached
0 pages in page table cache
Buffer memory: 184kB
Buffer heads: 244
Buffer blocks: 184
CLEAN: 114 buffers, 1 used (last=2), 0 locked, 0 protected, 0 dirty
LOCKED: 31 buffers, 20 used (last=24), 0 locked, 0 protected, 0 dirty
Networking buffers in use : 16
Total network buffer allocations : 1062
Total failed network buffer allocs : 0
IP fragment buffer size : 0
2:06pm up 12 min, 4 users, load average: 8.46, 7.12, 4.14
44 processes: 43 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: 0.3% user, 2.7% system, 0.2% nice, 97.3% idle
Mem: 4764K av, 4192K used, 572K free, 1500K shrd, 184K buff
Swap: 33228K av, 4684K used, 28544K free 788K cached
I had typed 'make bzImage' before the first memory output, it went nowhere.
After the 'echo "0 0"' it actually started compiling, albeit extremely
slowly.
2:12pm up 18 min, 4 users, load average: 8.44, 7.80, 5.38
50 processes: 49 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: 2.8% user, 5.4% system, 2.4% nice, 92.4% idle
Mem: 4764K av, 4192K used, 572K free, 984K shrd, 184K buff
Swap: 33228K av, 6272K used, 26956K free 708K cached
Mem-info:
Free pages: 520kB
( Free: 130 (48 96 144)
2*4kB 10*8kB 5*16kB 3*32kB 0*64kB 2*128kB = 520kB)
Swap cache: add 35609/35609, delete 35566/35566, find 0/0
Free swap: 26936kB
1536 pages of RAM
345 reserved pages
252 pages shared
43 pages swap cached
0 pages in page table cache
Buffer memory: 184kB
Buffer heads: 231
Buffer blocks: 184
CLEAN: 150 buffers, 1 used (last=2), 0 locked, 0 protected, 0 dirty
LOCKED: 28 buffers, 20 used (last=26), 0 locked, 0 protected, 0 dirty
DIRTY: 1 buffers, 0 used (last=0), 0 locked, 0 protected, 1 dirty
Networking buffers in use : 18
Total network buffer allocations : 3114
Total failed network buffer allocs : 0
IP fragment buffer size : 0
Still solid disk light, but the system is at least partially usable. I
sort of doubt any SMP machine would actually have 6 MB of RAM though. (This
is after a complete system boot, not just bash.)
PID USER RSS SHARE STAT %CPU %MEM TIME COMMAND
3 root 0 0 SW< 3.3 0.0 0:30 kswapd
The compile is currently stuck on sched.c. It doesn't seem to want to swap
out all of apache, even though it's not being used.
PID USER RSS SHARE STAT %CPU %MEM TIME COMMAND
208 nobody 52 52 S 0.0 1.0 0:00 httpd
209 nobody 52 52 S 0.0 1.0 0:00 httpd
210 nobody 52 52 S 0.0 1.0 0:00 httpd
211 nobody 52 52 S 0.0 1.0 0:00 httpd
212 nobody 52 52 S 0.0 1.0 0:00 httpd
echo "20 30 40" > /proc/sys/vm/freepages
echo "3 10 20" > /proc/sys/vm/buffermem
echo "10 30 50" > /proc/sys/vm/pagecache
Free pages: 172kB
( Free: 43 (20 30 40)
15*4kB 2*8kB 2*16kB 0*32kB 1*64kB 0*128kB = 172kB)
Swap cache: add 64273/64273, delete 64180/64180, find 0/0
Free swap: 26444kB
1536 pages of RAM
345 reserved pages
269 pages shared
93 pages swap cached
0 pages in page table cache
Buffer memory: 144kB
Buffer heads: 180
Buffer blocks: 144
CLEAN: 101 buffers, 1 used (last=1), 0 locked, 0 protected, 0 dirty
LOCKED: 20 buffers, 18 used (last=20), 1 locked, 0 protected, 0 dirty
DIRTY: 3 buffers, 3 used (last=3), 0 locked, 0 protected, 3 dirty
Networking buffers in use : 841
Total network buffer allocations : 7372
Total failed network buffer allocs : 0
IP fragment buffer size : 0
The pages free looks a lot more reasonable now. Hm, now the network
buffers are even higher: (only network activity being a telnet session to
another Linux machine, 2.0.33)
Mem-info:
Free pages: 140kB
( Free: 35 (20 30 40)
5*4kB 3*8kB 2*16kB 0*32kB 1*64kB 0*128kB = 140kB)
Swap cache: add 67611/67611, delete 67489/67489, find 0/0
Free swap: 26056kB
1536 pages of RAM
345 reserved pages
210 pages shared
122 pages swap cached
0 pages in page table cache
Buffer memory: 140kB
Buffer heads: 186
Buffer blocks: 140
CLEAN: 90 buffers, 1 used (last=1), 0 locked, 0 protected, 0 dirty
LOCKED: 18 buffers, 18 used (last=18), 1 locked, 0 protected, 0 dirty
DIRTY: 3 buffers, 3 used (last=3), 0 locked, 0 protected, 3 dirty
Networking buffers in use : 1020
Total network buffer allocations : 9392
Total failed network buffer allocs : 0
IP fragment buffer size : 0
Odd.
-George
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html