On Mon, 2010-05-31 at 19:59 +0200, Philippe Gerum wrote: > On Mon, 2010-05-31 at 19:05 +0200, Jan Kiszka wrote: > > Hi Philippe, > > > > something is fishy in I-pipe for 2.6.34 (both w/ and w/o Xenomai, > > goes away once I-pipe is off): > > > > top - 18:55:26 up 3 min, 2 users, load average: 2.09, 1.42, 0.58 > > Tasks: 60 total, 1 running, 59 sleeping, 0 stopped, 0 zombie > > Cpu(s): 0.0%us, 24.7%sy, 0.0%ni, 75.3%id, 0.0%wa, 0.0%hi, 0.0%si, > > 0.0%st > > Mem: 1012496k total, 121264k used, 891232k free, 4676k buffers > > Swap: 1156644k total, 0k used, 1156644k free, 80748k cached > > > > PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND > > 5636 root 20 0 51776 3.7t 3.7t S 0.0 392333.0 0:00.01 sshd > > 4195 haldaemo 20 0 34220 1.8t 1.8t S 0.0 188901.4 0:02.10 hald > > 5327 root 20 0 19036 1.5t 1.5t S 0.0 159839.4 0:00.15 syslog-ng > > 5495 root 20 0 164m 1.4t 1.4t S 0.0 145308.6 0:00.23 nscd > > 5563 root 20 0 16240 1.4t 1.4t S 0.0 145308.6 0:00.01 smartd > > 5523 root 20 0 10204 982g 982g S 0.0 101716.0 0:00.01 > > avahi-dnsconfd > > 5330 root 20 0 4012 701g 701g S 0.0 72654.3 0:00.04 klogd > > 4154 root 20 0 3872 561g 561g S 0.0 58123.4 0:00.04 acpid > > 5991 root 20 0 13264 2356 1596 S 0.0 0.2 0:00.34 bash > > > > I swear this machine does *not* yet have Tera bytes of RAM (just 1G). > > > > Looks like /proc/<pid>/stat is broken, likely even something deeper. > > Philippe, did you ran into suspicious conflicts in the mm subsystem > > while rebasing over this kernel? Any spontaneous idea? > > Bad hunks in mm/memory.c.
This might even date back to 2.6.33, with changes in the vm accounting IIRC. > > > > > Jan > > > > -- Philippe. _______________________________________________ Adeos-main mailing list [email protected] https://mail.gna.org/listinfo/adeos-main
